Journal of Child Psychology and Psychiatry 52:4 (2011), pp 368–397 doi:10.1111/j.1469-7610.2010.02306.x

Annual Research Review: All mothers are not created equal: neural and psychobiological perspectives on mothering and the importance of individual differences

Jennifer Barrett and Alison S. Fleming Department of Psychology, University of Toronto at Mississauga, Mississauga, Ontario, Canada

Quality of mothering relies on the integrity of multiple physiological and behavioral systems and on two maternal factors, one proximal and one distal, that have a great impact on how a mother mothers: postpartum depression and early experiences. To mother appropriately requires the action of systems that regulate sensation, perception, affect, reward, executive function, motor output and learning. When a mother is at risk to engage in less than optimal mothering, such as when she is depressed or has experienced adversity in childhood, the function of many or all of maternal and related systems may be affected. In this paper, we will review what is currently known about the biological basis of mothering, with attention to literature on hormones but with a particular focus on recent advances in the fields of functional neuroimaging. Instead of discussing strictly ‘maternal’ brain imaging studies, we instead use a systems approach to survey important findings relevant to brain systems integral to and/or strongly related to the mothering experience: (a) social behavior; (b) reward and affect; (c) executive function; and (d) maternal behavior. We find that there are many commonalities in terms of the brain regions iden- tified across these systems and, as we would expect, all are sensitive to the influence of, or function differently in the context of, depression and adverse early experience. It is likely that the similarity and cross-talk between maternal, affect and stress systems, observed behaviorally, hormonally and in the context of brain function, allows for mood disturbance and early adverse experiences to have a signif- icant impact on the quality of mothering and the motivation to mother. Keywords: Adversity, maternal depression, brain imaging, parenting, mothers.

The interaction between a mother and her infant can baby’s needs, that she learns from her experiences be like a dance. There are routines, standards and and that she behaves appropriately. missteps, there is give and take, there is unparalleled In humans, across cultures there exist as many intimacy, there are often vast differences in skill level differences as similarities in the ways that mothers and motivation, there is learning. A mother must interact with their infants (Leiderman & Leiderman, learn to be an adept partner, being sensitive to the 1977). In some cultures the infants are carried in needs of her offspring while ultimately guiding the cradleboards on the mothers’ backs; in others they quality and nature of care to ensure normal growth are carried in a pouch or sling on the mothers’ and development. This dance can beautiful, it can be ventrum; in many Western cultures, especially if tender, it can be awkward, it can be difficult. And the terrain is smooth, infants are transported in sometimes it just does not occur! strollers or perambulators (Anisfeld, Casper, Nozyce, Romanticism aside, the mother–offspring dance is & Cunningham, 1990; Chisholm, 1978). Customs influenced by and interacts with many factors, relating to sleeping also vary across cultures, from including mothers’ physiology, cultural and family swaddling, sleeping in the mother’s bed, to sleeping context, maternal cognitions, maternal affect and in a cot next to the mother, to sleeping away from the stress and the early environment, notably a mother’s mother in another room (Van Sleuwen et al., 2007; own early experiences in her family of origin. Bio- Thoman, 2006). Mothers also exhibit a range of dif- logical influences certainly include how mothers’ ferent kinds of communicative responses to their brains are organized and how genes and environ- infants: some look at them directly while others gaze ment interact in brain development. It is believed avert (Brazelton, 1977); some keep their babies that over the course of pregnancy and through the unclothed and stroke their bodies; others swaddle postpartum period, changes in the levels of certain them instead (see Corter & Fleming, 2002). Some hormones likely ‘activate’ or affect key brain regions talk or sing to their babies; others do not (Tronick, to ensure that the mother is especially attracted to 1987). Soothing methods, for example pacifier use or her baby, that she is attentive and sensitive to her use of specific herbs or teas, can greatly vary by culture (Abdulrazzaq, Kendi, & Nagelkerke, 2009). More subtly, within a culture mothers show Conflict of interest statement: No conflicts declared. large variations in the postpartum development of 2010 The Authors. Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. Published by Blackwell Publishing, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main St, Malden, MA 02148, USA Neurobiological perspectives on mothering 369 nurturant feelings, from minutes to months (Leifer, output and learning. When a mother is at risk to 1980; Moss & Jones, 1977; Robson, 1967; Robson & engage in less than optimal mothering, such as when Kumar, 1980; Trevathan, 1983), and, once ‘attached’ she is depressed or has experienced adversity in or emotionally committed, in the intensity with childhood, the function of many or all of these sys- which they exhibit different caregiving behaviors. tems may be affected. In this paper, we will use a Within all cultures most mothers are motivated to systems approach to review and critique what is provide warmth, shelter, and food to the infant while currently known about the biological basis of moth- in all cultures there are mothers who neglect or even ering, with a particular focus on recent advances in abuse their infants (see Corter & Fleming, 2002; the field of functional neuroimaging. We will often Hrdy, 2005, 2009). combine information from the animal world with the Despite these differences, in most cultures moth- human. We will briefly review what has informed ers nurse their infants, although there are some knowledge of the neurobiology of mothering up until mothers in each that may also bottle-feed (Kannan, this point (hormones). Finally, whenever possible, we Carruth, & Skinner, 2004; Pak-Gorstein, Hak, & will try to put these basic mechanisms into the Graham, 2009; Thulier, 2009). In many different context of important individual differences related to cultures, mothers engage in a form of expression maternal psychopathology (depression) and early called ‘motherese’, where the mother talks to the adversity. infant in a high prolonged pitch and sings with a characteristic cadence and rhythm (Trehub, Unyk, & Trainor, 1993). In most cultures mothers provide I. Maternal behavior is affected by mothers’ kinesthetic stimulation by stroking and rocking the cognitions, executive function, and theory of baby and respond contingently to infant cries and mind other indices of distress, often by vocalizing ‘in Overview response’ (Ainsworth, Bell, & Stayton, 1974; Corter & Fleming, 2002; Pederson et al., 1990). Here we will How mothers relate to their infants is affected by not address the issue of cross-cultural differences, multiple psychological propensities, including her important though they are (Cote & Bornstein, 2009; cognitive style and her affective state. In this first Gu¨ngo¨r & Bornstein, in press). Instead we focus on section we discuss individual differences related the cross-cultural similarities and focus most dis- to cognitive functions: functions that are thought to cussion on studies that describe mothers from provide an important foundation for quality of Western and industrialized societies, where most of mothering and maternal behavior and are also, in the psychobiological research has occurred (see the many cases, known to interact with affect/mood and work of Marc Bornstein for a program of research early experiences. Processes such as cognitive flexi- well suited to a cross-cultural analysis). bility, working memory and attentional control are For this paper our primary focus is reflected in the fundamental components of mothering and key to idea that mothering is not unitary, but is instead maternal sensitivity. A mother must have the atten- complex, and comprised of many behavioral pro- tional command to focus on her infant which allows pensities that are mediated, moderated, and regu- her to be sensitive to infant needs, that is, to respond lated by multiple physiological and brain systems. contingently, appropriately and in a timely manner. To engage in mothering behavior, mothers have to be She must also have the cognitive flexibility to switch sensitive to infant cues and select those cues for her attention efficiently across many situational processing, utilizing multiple sensory and percep- demands, in highly stimulating environments. She tual modalities; the cues must be attractive and must also maintain and manipulate information in salient for the mother, recruiting reward and her working memory to plan and guide mother– approach systems. Mothers must be emotionally infant interaction and infant care. In addition, the prepared and positively motivated to engage socially application of the concept of Theory of Mind (ToM) to with the infant, depending on systems regulating mothering, where the strongly related concept of affect. They must selectively attend to the offspring ‘mind-mindedness’ has also been examined (see in the context of competing stimuli, enacted through Meins et al., 2003), is relatively new and exciting. systems that regulate attention; and they must be Theory of Mind, which, among other things, is a restrained and consistent in their responsiveness, theory relevant to social cognition and describes how depending on systems that regulate impulsivity. we understand the intentions and emotions of oth- Finally, mothers gain through experiences, acquired ers, has obvious implications for mothering, where both early in life and with young as juveniles and in correctly identifying and responding to infant signals adulthood. These experiences are acquired, consoli- or intentions is key. dated, and stored as motor or sensory memories and are based on extensive brain plasticity. Executive function Thus, to mother appropriately requires the action of multiple systems in the domains of sensation, Work in the rat has demonstrated that the integrity perception, affect, reward, executive function, motor of maternal behavior is linked to executive functions

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. 370 Jennifer Barrett and Alison S. Fleming such as attentional set-shifting and prepulse inhi- (Burton, Lovic, & Fleming, 2006; Garner, Wood, bition (e.g., Afonso, Sison, Lovic, & Fleming, 2007; Pantelis, & van den Buuse, 2007; Lovic & Fleming, Lovic & Fleming, 2004). In terms of human mothers, 2004). Many studies of early adversity in children recent work in our lab by Andrea Gonzalez examined have found a relationship between abuse or neglect neuropsychological and physiological factors as and executive function (Bos, Fox, Zeanah, & Nelson, intervening variables between maternal retrospective 2009; De Bellis, 2005; Kreppner, O’Connor, & reports of early adversity (childhood maltreatment Rutter, 2001; Pears & Fisher, 2005). Bos and col- and consistency of care) and current parental ability leagues examined executive function (CANTAB) in (maternal sensitivity) in a sample of 90 mothers. In children with a history of early deprivation as a this work, Gonzalez proposed and tested a model result of institutionalization and found that early linking maternal early adversity to maternal sensi- adversity is related to performance deficits on tasks tivity through the hypothalamic-pituitary-adrenal involving planning and working memory (Bos et al., (HPA) axis and executive functions (e.g., attention). It 2009). Early adversity as a result of familial violence is of note that the latter was predicted to be impor- has also been linked to deficits in a broad range of tant for maternal behavior and thought to be medi- executive functions, including problem solving, ated by HPA axis function. A composite maternal working memory, inhibition and attentional control sensitivity score was correlated with measures of (Fishbein et al., 2009; Nolin & Ethier, 2007; Pears, attentional set-shifting, spatial working memory, Kim, & Fisher, 2008). Although we may expect that and a sustained attention measure; results revealed observed early adversity-related deficits in executive that maternal sensitivity was significantly related to function may persist into adulthood, less is known both set-shifting and spatial-working memory. about the general influence of early adversity on Importantly, path analyses revealed that maternal executive function in adults. In the context of mood, early adversity was indirectly related to maternal deficits in executive function have also been widely sensitivity via two pathways: (1) early adversity–HPA linked to major depression and other affective dis- function–maternal sensitivity; and (2) early adver- orders (Fossati, Ergis, & Allilaire, 2001). Although sity–HPA function–executive function–maternal results have been mixed (see Rogers et al., 2004), sensitivity. In this population of mothers, the moth- individuals with depression have been shown to have ers’ affective state did not enter into the model. Of difficulties with executive functions such as working interest to us is the possibility that in a population of memory (Elliott et al., 1996; Kaneda, 2009; Sweeney, mothers where postpartum depression is high, early Kmiec, & Kupfer, 2000), set-shifting (Grant, Thase, experience effects may also affect cognition and & Sweeney, 2001; Moritz et al., 2002) and atten- maternal sensitivity by acting to alter mothers’ tional control (Nakano et al., 2008). affective state. However, the importance of HPA function in mothering will be described in detail, in Theory of Mind the section to follow. Atkinson and colleagues investigated mother– Maternal mind-mindedness reflects the tendency to infant interaction, attachment and selective atten- treat one’s infant as an autonomous human being tion using the emotional Stroop task and found that with complex desires and intentions and not just mothers with a classification of disorganized basic or reflexive organismic needs, as well as the attachment responded more slowly to negative ability to correctly interpret infants’ implicit and attachment words and that the speed of response to explicit signals (Meins et al., 2003). Interestingly, such stimuli was correlated with the number of this trait or tendency has been identified across the times the mother–infant dyad was labeled as disor- perinatal period, even before the infant is born, with ganized (Atkinson et al., 2009). The authors interpret mothers more likely to comment on the ‘personality’ these findings in the context of parallel processing of the fetus exhibiting more appropriate comments and distributed semantic (neural) networks, where regarding the infant’s internal state during mother– attachment stimuli (i.e., infant) are connected to infant interaction at six months postpartum (Arnott negative associations (e.g., threat, loss). Such & Meins, 2007). Individual differences in mind- mechanisms may contribute to ongoing cognitive mindedness have been associated with maternal (attentional) interference/difficulties during mother– sensitivity. The work of Meins and colleagues (1998, infant interactions (Atkinson et al., 2009). The 2001) has demonstrated that more securely attached results of these studies underscore the complexity of mothers are more likely than insecurely attached the mechanisms underlying mothering and how a mothers to have high mind-mindedness and focus multi-system approach is important when trying to on their children’s mental or cognitive attributes, understand individual differences that may contrib- rather than their physical appearance or behavioral ute to maladaptive mothering styles. tendencies. More recently, Laranjo, Bernier, and In terms of individual differences in mothering and Meins (2008) examined the relationship between executive functions, in rats maternal deprivation is maternal sensitivity, maternal mind-mindedness related to deficits in executive function-processes and infant attachment: 50 mother–infant dyads were such as prepulse inhibition (PPI) and set-shifting assessed at two time points (15 and 18 months).

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. Neurobiological perspectives on mothering 371

Results revealed that sensitivity, attachment, and depression usually remits by 6 months postpartum, mind-mindedness were all positively interrelated. In it is a source of considerable distress for the mother addition, the relationship between mind-minded- and her family. A great challenge of any type of ness and attachment was partly mediated by depression experienced in the postpartum period is maternal sensitivity, a finding that was thought to that it is in direct conflict with societal and often self- support the notion that mind-mindedness is a pre- expectations of what being a new mother should and requisite for maternal sensitivity, and that the cor- will feel like. The truth is, the early postpartum rect interpretation of infant cues is fundamental to period is a time of upheaval in the physiology and the generation of an appropriate response to such psychology of the mother, when in a short time she cues (Laranjo et al., 2008). evolves from being a pregnant women to the care- Interestingly, one individual difference that has giver of a newborn, and this change comes with new been associated with Theory of Mind abilities is non- and considerable responsibilities. Although the pre- postpartum depression. Theory of Mind deficits have cise reasons for it are unclear, mood disturbances been found in remitted depressed patients and been during this time are incredibly common and many demonstrated to be risk factors for depression new mothers feel depressed at some point after the relapse (Inoue, Tonooka, Yamada, & Kanba, 2004; birth of their infants (Friedman & Resnick, 2009; Inoue, Yamada, & Kanba, 2006). In the context of Marcus, 2009; Moses-Kolko & Roth, 2004). Mild depression, Theory of Mind deficits have been asso- depressed mood, often termed ‘baby blues’, is expe- ciated with impaired social cognition and cognitive rienced by up to 85% of mothers and typically distortions regarding the state of mind of others. resolves without treatment by the end of the first Such impairments in social functioning may con- postpartum month (Buist, 2006; O’Hara, Zekoski, tribute to or help perpetuate the biased, negative Philipps, & Wright, 1990). thinking and false beliefs that often characterize depressive episodes (Sheline et al., 2009). Relatedly, Postpartum depression one may expect that infant-related social cognition deficits related to Theory of Mind may play a strong In some women, mood disturbance may persist role in depression during the postpartum period, beyond the initial postpartum period, leading to where depressed mothers may be more likely to more serious postpartum depression. Epidemiologic incorrectly assess the intentions of their infants and studies have found that up to 20% of women exhibit be biased toward more negative assessment (e.g., my symptoms of depression in the first weeks following baby is crying because she hates me). delivery (Gavin et al., 2005; Marcus, 2009), that the rise in incidence is within 30 days post-partum but persists for up to two years, and that the great II. Maternal behavior is affected by concurrent majority of these depressive episodes resolve spon- mood or depression taneously within six months to a year (Cooper & Murray, 1995; Cox, Murray, & Chapman, 1993). The Overview symptom profile of postpartum depression, which One of the most profound experiences a woman can includes sad mood, restlessness/agitation, and have is giving birth to a new baby. She experiences impaired concentration, resembles that of a major intense emotions involving both extreme elation and depressive episode experienced at other times in life, happy feelings mixed with mild distress, lability, and but it is unique in its timing and usually involves the tearfulness. Rapid changes in affect and extremes in mother–baby dyad and in most cases affects the affect intensity are very common and may function to entire family unit. Interestingly, there is also some heighten mothers’ earliest impressions and experi- suggestion that greater psychomotor disturbance ences of the baby. Likely hormonally mediated, the may distinguish postpartum from non-postpartum first postpartum week is for most women an exciting depression (Bernstein et al., 2008). As with depres- and positive experience and one that maps on to sion outside of the perinatal period, postpartum many changes in mothers’ awareness of their new depression often co-occurs with other psychiatric role and the experience of ‘falling-in-love’ with the disorders, the most common being generalized anx- baby, with all the anxiety and happiness that that iety disorder, obsessive-compulsive disorder, social experience normally entails. Fleming and colleagues phobia and agoraphobia (Figueira, Fernandes Mal- describe the development of these strong positive loy-Diniz, Aure´lio Romano-Silva, Silva Neves, & feelings of nurturance – and the associated anxiety – Correˆa, 2009). In some rare cases, postpartum as they occur across pregnancy and through the first depression has been linked to suicidality and filicide postpartum year (Fleming et al., 1997a). (Kauppi, Kumpulainen, Vanamo, Merikanto, & In contrast to these positive feelings are the Karkola, 2008; Paris, Bolton, & Weinberg, 2009). more negative feelings experienced during the first Alarmingly, studies have also shown that as many as postpartum months by a substantial number of 50% of postpartum depression cases are undiag- women (Friedman & Resnick, 2009; Gale & Harlow, nosed (Chaudron et al., 2005; Murray, Woolgar, 2003; Marcus, 2009). Although this dysphoria or Murray, & Cooper, 2003), and reasons for this

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. 372 Jennifer Barrett and Alison S. Fleming statistic include lack of attention on the mental Field, Diego, & Hernandez-Reif, 2009; Field, Diego, health of the mother by health care providers as Hernandez-Reif, Schanberg, & Kuhn, 2002; Field well as under-reporting of symptoms by the mother et al., 2007) and is associated with increased risk of herself. Many factors have been found to influence developing psychopathology in adolescence or young or mediate mothers’ vulnerability to becoming adulthood (Brand & Brennan, 2009; Cicchetti, Rog- depressed in the postpartum period, including: (1) osch, & Toth, 1998; Gump et al., 2009; Halligan, factors related to childbirth and childcare such as Murray, Martins, & Cooper, 2007). sleep patterns, type of delivery and lack of social support; and (2) non-specific factors such as history of depression outside of the perinatal period, seasonal III. Maternal behavior is affected by early mood fluctuations, socioeconomic status, lack of experiences in family of origin social support, stressful recent life events and marital Overview conflict (Boyce & Todd, 1992; Goyal, Gay, & Lee, 2009; Koo, Lynch, & Cooper, 2003; Panthangi, West, As we have seen, how mothers feel at the time they Savoy-Moore, Geeta, & Reickert, 2009; Vik et al., are mothers, whether depressed, anxious or stres- 2009; Xie, He, Koszycki, Walker, & Wen, 2009). sed, has an impact on the quality and intensity of their maternal responsiveness. These affective characteristics that exert a proximal influence on Mothering and maternal mood mothering behavior have their own origins, often Mothers with postpartum depression tend to be more within the mothers’ early experiences. In fact, among intrusive and irritated, respond less sensitively, the most powerful influences on how mothers’ contingently and more negatively to their infants and mother their infants come from their own experi- demonstrate disrupted patterns of communication ences growing up. Important aspects of mothering in compared with mothers without postpartum the child’s early life include the quality of the inter- depression (Beebe et al., 2008; Cohn, Campbell, action, the warmth expressed by the mother, Matias, & Hopkins, 1990; Field, Healy, Goldstein, & whether the mother–infant interaction involves pre- Guthertz 1990; Field et al., 1985; Fleming, Ruble, dominantly physical contact and play, vocal Flett, & Shaul, 1988; Paris et al., 2009; Righetti- exchanges, and/or visual mutuality. Also important Veltema, Conne-Perre´ard, Bousquet, & Manzano, for mothering is the extent to which mothers respond 2002, for a review see Field, 2009). When observed contingently to their infants, regardless of modality later in the postpartum period, depressed dyads (i.e., (Bornstein, Tamis-Lemonda, Hahn, & Haynes, 2008; mother–baby pairs) exhibit reduced mutual atten- Tamis-LeMonda, Bornstein, & Baumwell, 2001). tiveness, vocal and visual communications, touching Contingent responding encompasses the concept of interactions or smiling compared with non- mutuality and is one aspect of mothering that postpartum depression dyads (Field et al., 1990; ‘empowers’ the developing infant by providing an Fleming et al., 1988; Righetti-Veltema et al., 2002). experience of efficacy (for an example see Tamis- For example, a recent study by Field and colleagues LeMonda & Bornstein, 1989). Hence, parents pro- (2007) examined mother–infant interaction in vide experiences through expression of acceptance, depressed and non-depressed dyads using the still- affection, liking and caring. face paradigm. Infants were approximately 4 months Parents also provide ‘negative’ and adverse expe- old at the time of testing. One of the authors’ main riences that impact infants and the way they come to findings, in terms of maternal responsiveness, was parent their own infants when they grow up. Adverse that the mothers with postpartum depression were early experiences or early adversity has been opera- less interactive with their infants at baseline (less tionalized or quantified in a number of different smiling and moving of infant limbs) as well as in the ways, including socioeconomic factors related to still-face reunion period (less smiling, touching and housing and parental employment, illness affecting moving of infant limbs). The latter finding is espe- the child or close family members, family discord cially telling as a main feature of the reunion period including parental separation and divorce, and of the still-face paradigm is to attempt to reinstate finally, parenting practices and parenting styles more positive interaction or regulate the emotion of (Bifulco et al., 2006; Carpenter et al., 2004; the infant. It is of note that more practical, but Karevold, Røysamb, Ystrom, & Mathiesen, 2009; equally important, maternal activities are also Klein et al., 2009; Levitan, Rector, Sheldon, & Goe- influenced by depression, including breastfeeding, ring, 2003; Nachmias, Gunnar, Mangelsdorf, Parritz, bedtime routines, medical care and safety practices & Buss, 1996; Rao, Hammen, Ortiz, Chen, & Poland, (reviewed in Field, 2009). 2008; Repetti, Taylor, & Seeman, 2002; Rikhye et al., Postpartum depression also has deleterious impli- 2008). Here we will focus on early experience and cations for the cognitive, emotional, motor and even early adversity as they relate to parenting practices neural functioning of the infant (Tronick & Reck, and styles. These include physical, sexual and 2009; also see Diego, Jones, & Field, 2010; Diego emotional abuse, parental neglect in terms of disin- et al., 2004; Diego et al., 2006; Feldman et al., 2009; terest in material care and health, direct and indirect

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. Neurobiological perspectives on mothering 373 control exerted over the child by the parent, and spring (Champagne & Meaney, 2001). Thus, as we parental psychopathology such as depression and would expect, maternal behavior deficits observed in anxiety. Other known influences on mothers’ ability adulthood are related to the degree of maternal to mother include the stability of their early family deprivation received as a neonate (Gonzalez, Lovic, environment, the presence of both their primary Ward, Wainwright, & Fleming, 2001; Lovic et al., caregivers as well as other immediate family mem- 2001; Rees, Akbari, Steiner, & Fleming, 2008) and, bers (e.g., siblings). In general, these early experi- importantly, observed behavioral differences are ences that mothers have been exposed to affect the accompanied by structural brain changes (Burton et subsequent attachment status of the mother–child al., 2007; Chatterjee et al., 2007; Gonzalez & Flem- dyad, reflected in specific maternal behaviors and ing, 2002). Offspring of high licking moms become mothering styles. high licking moms (Champagne & Meaney, 2001). Rat pups reared in isolation from their mother and without such stimulation, however, show maternal Early experiences and intergenerational transfer of behavior deficits in behaviors such as crouching and mothering styles licking (Gonzalez et al., 2001; Melo et al., 2006), an There is a growing literature both in humans and in effect that is retained even towards foster pups other animals that shows that the quality of mother- (Palombo, Nowoslawski, & Fleming, 2010). Further- ing received from one’s own mother influences the more, deficits are related to the degree of early quality of mothering the daughter expresses with her maternal separation with more extreme separation own offspring. For example, in rats, maternal traits paradigms associated with the most severe behav- such as licking and grooming have been shown to be ioral deficits (Lovic et al., 2001; Rees et al., 2008). transmitted across generations (reviewed in Cham- Maternal separation and the associated reductions pagne, Francis, Mar, & Meaney, 2003; Fleming et al., of stimulation in pups are further related to the 2002; Numan, Fleming, & Levy, 2006). In human progressive dysregulation of physiological processes mothers, evidence exists for the intergenerational in the offspring as well as behavioral deficits related transmission of affective traits (Miller, Warner, to social learning, sustained attention, inhibition of Wickramaratne, & Weissman, 1999), parenting styles acoustic startle, and perseveration. It is generally (Putallaz, Costanzo, Grimes, & Sherman, 1998; Ser- accepted that early experience, notably the quality bin & Karp, 2004; van IJzendoorn, 1992), as well as and quantity of maternal care, influences the plas- maternal factors such as bonding and attachment ticity and function of set of brain structures that are (Benoit & Parker, 1994; Miller, Kramer, Warner, genetically mandated to express maternal behavior; Wickramaratne, & Weissman, 1997; van IJzendoorn, this circuit interfaces with brain structures that 1992). mediate emotion, sensation, perception, memory How parenting is transmitted across generations and cognitive functions. and the importance of early experience in the family In human mothers, research also suggests that of origin are just really beginning to be understood, early adversity as indexed by quality of care in the mostly at the cognitive or behavioral level in human family of origin may also affect subsequent parenting moms. There is, however, a large body of work on the (Belsky, 1984; Champagne & Meaney, 2001; Flem- behavioral and biological bases of the intergenera- ing et al., 2002; Maestripieri, 2005). The significance tional transmission of maternal style in rat mothers of this statement is underscored by the fact that and as this work is relevant to the present discussion maternal history of abuse has been proposed as a we will now describe it in brief. Four major compo- risk factor for child maltreatment. Approximately nents of maternal behavior in rats are retrieval 30% of mothers who were abused as children go on behavior, nest building, nursing behavior, and pup to abuse their own children, compared to 5% in grooming, which includes licking (Numan & Insel, mothers not reporting abuse (Knutson, 1995). While 2003). In rat studies of mothering, the best approx- the mechanisms underlying a posited ‘cycle of abuse’ imation of variation in maternal care is demon- are not well understood, as previously noted, there is strated by the elegant studies of Michael Meaney and support for the intergenerational transmission of his students in which they show that the licking parenting styles, which includes factors such as exhibited by mother rats shows a very large varia- bonding, attachment and maternal rejection (Benoit tion, variation that influences the quality and & Parker, 1994; Miller et al., 1997; Serbin & Karp, intensity of the licking behavior of their female off- 2004; van IJzendoorn, 1992). Moehler, Biringen, and spring that become the next generation of mothers Poustka (2007) analyzed whether emotional avail- (Champagne & Meaney, 2001). The observation from ability measured during mother–infant interaction is cross-fostering studies that the licking profile of off- altered by experiences of physical or sexual abuse in spring is determined by their preweaning experi- the mother’s family of origin. Results revealed that ences (hence, the pattern shown by the foster mothers with a history of physical or sexual abuse mothers) and not by their genetic origins suggests were significantly more intrusive toward their chil- that the postnatal experiences of the young are dren than were control mothers. The relationship crucial to their later behavior with their own off- between maternal early experiences and perceived

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. 374 Jennifer Barrett and Alison S. Fleming parental stress has also been examined. In a study of effects or cause development to follow a different 120 mothers, Willinger and co-authors (Willinger, path (McEwen, 2003). In support for this, recent Diendorfer-Radner, Willnauer, Jo¨rgl, & Hager, 2005) work by Bergman, Sarkar, Glover, and O’Connor (in measured recalled parental bonding and parenting press) demonstrated that the (potentially) deleteri- stress and found that higher levels of parenting ous influence of maternal prenatal cortisol on infant stress were experienced by mothers reporting less development is mediated by mother–infant attach- optimal parental bonding. Adverse early experiences ment in the second year of life, with more securely are also risk factors for difficulties in mothering, attached toddlers showing no effect of prenatal albeit indirectly, through their relationship with maternal cortisol on cognitive abilities and inse- maternal psychopathology. curely attached toddlers showing a significant, Epidemiological studies indicate that children positive relationship. exposed to early adverse experiences, including those related to bonding and attachment, are at increased risk for the development of depression, anxiety and IV. What has informed us up until this point: other stress-related disorders in adulthood (Enns, Hormones Cox, & Clara, 2002; Putnam, 2003) and during ado- Basic mechanisms lescence (Herrenkohl et al., 2010; Martin, Bergen, Roeger, & Allison, 2004; Rey, 1995). Spertus, Yehuda, Hormones play a fundamental role in maternal Wong, Halligan, and Seremetis (2003) found that a behavior and have been a focus of much research in history of emotional abuse and neglect was associated animal mothers and only more recently, in humans. with increased anxiety, depression, posttraumatic Research in rats and other mammals has shown that stress and physical symptoms and that long-standing the hormonal cocktail of pregnancy and parturition, behavioral consequences may arise as a result of which includes high levels of the peptide hormones, childhood emotional abuse and neglect, specifically, oxytocin and prolactin, against a background of poorer emotional and physical functioning, factors changing levels of steroid hormones, estradiol and that would certainly put these women at risk for diffi- progesterone, is essential for the onset and mainte- culties in mothering. In addition, an intergenerational nance of maternal behavior (Bridges, 1990; Bridges, effect may account for cyclical occurrence of teenage 2008; Insel, 1990; Numan et al., 2006; Pryce, Mar- motherhood (de Pau´ l & Domenech, 2000; Rebollo & tin, & Skuse, 1995; Rosenblatt, Olufowobi, & Siegel, Montero, 2000; Giardino, Gonzalez, Steiner, & Flem- 1998). ing, 2008), where teen mothers often raise poorly Although evidence for this is stronger in rat com- adjusted children who tend to become teen parents pared with human mothers, this hormonal profile is (Furstenberg, Brooks-Gunn, & Morgan, 1987). thought to increase mothers’ attraction to infant Now a disclaimer! Despite these clear relations cues, enhance the reinforcing value of pups, and between early adversity (and/or retrospective per- influence affective state (Fleming, Ruble, Krieger, & ception of that adversity) and later emotional and Wong, 1997a; Numan et al., 2006; Numan & Insel, parenting problems, the statistic in humans relating 2003). In pregnant women, feelings of attachment to to the intergenerational transmission of abuse the fetus appear to be unrelated to changing levels of indicates that 70% of mothers who were abused will pregnancy hormones. However, Fleming et al. NOT abuse their own offspring, indicating that there (1997a) did find a relationship between hormones exists a host of factors that protect prospective and attachment: high attachment to new babies mothers from the riskiness of their early experiences. postpartum was related to an increase from early to Included among the buffers that are in play are late pregnancy in estradiol/progesterone ratio and having social supports or a supportive adult some- low attachment was related to a decrease in the where in the environment of the developing child estradiol/progesterone ratio. (Jaffee, Caspi, Moffitt, Polo-Toma´s, & Taylor, 2007; Furthermore, mothers with a greater shift in the Kaufman et al., 2004; Wind & Silvern, 1994) as well estradiol to progesterone ratio across pregnancy also as forming a partner relationship with a supportive experienced greater feelings of well-being in the spouse in adulthood (Seeman, Singer, Ryff, Dienberg postpartum period. Hormones and well-being Love, & Levy-Storms, 2002). These buffers also together explained 40% to 50% of the variance in speak to the issue of the plasticity of the mothering mothers’ attachment (Fleming et al., 1997a). A role system. Early experiences can have profound influ- for oxytocin in humans is also suggested by recent ences on the development of all the constituents that work by Feldman, Weller, Zagoory-Sharon, and must develop for appropriate mothering to be Levine (2007) in which they found that oxytocin expressed in terms of affect, attention, learning and levels both during early pregnancy and postpartum so on. And it may well be that the early periods are were associated with behaviors in mothers normally more susceptible to ‘negative’ experiences than are measured in mother–infant interactions, including the later periods in development (Heim & Nemeroff, gaze, voclizations, positive affect, and affectionate 1999; McEwen, 2003); however, experiences touch; they also found oxytocin levels related to acquired later on are either able to reverse earlier attachment-related thoughts. These results are

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. Neurobiological perspectives on mothering 375 striking in their similarity to what has been frequently observed in major depression. Depression described in other mammals and likely reflect an is thought to be characterized by reduced sensitivity effect of oxytocin in a mother who has been previ- of stress-hormone-related negative feedback mech- ously primed by estrogen. anisms which are associated with larger hormonal Similar to what has been shown in rat mothers response to stressors, higher baseline levels of stress (Rees, Panesar, Steiner, & Fleming, 2004, 2006), hormones and reduced suppression of the HPA axis postpartum hormones from the hypothalamic- in response to the dexamethasone test (Heim et al., pituitary-adrenal (HPA) axis may also be important 2004). Studies of corticotrophin releasing factor in mothers’ response to their newborns. The HPA measured in human lumbar cerebrospinal fluid axis and HPA-reactivity have been widely studied in (CSF) have shown increased levels of this stress- humans and animals in the context of emotion, related peptide in depressed patients compared to stress and psychopathology (see Heim & Nemeroff, control participants (Banki, Bissette, Arato, O’Con- 1999; Heim, Plotsky, & Nemeroff, 2004; McEwen, nor, & Nemeroff, 1987; Banki, Karmacsi, Bissette, & 2003). We have (Fleming, Steiner, & Anderson, 1987; Nemeroff, 1992; France et al., 1988; Nemeroff et al., Corter & Fleming, 1990) examined cortisol in rela- 1984). When examined post-mortem, the brains of tion to maternal behavior in the early postpartum depressed patients have been found to have in- period when cortisol levels are relatively high and creased CRF mRNA and increased CRF concentra- mothers’ emotional status is labile. Higher cortisol tions in the hypothalamus. In addition, CRF levels in levels on days 3 and 4 postpartum were significantly CSF have been found to normalize following suc- and strongly associated with maternal approach cessful antidepressant treatment (Nemeroff, Bis- behaviors, positive maternal attitudes, or more sette, Akil, & Fink, 1991; Rudorfer, Risby, Osman, vocally active infants. Generally, our work suggests Gold, & Potter, 1991). It is important to note that, as that in the early postpartum period (i.e., first post- with postpartum depression, results for HPA axis partum week) in first-time mothers, the HPA system function in depression have been mixed; one may influence maternal responsiveness by enhanc- intriguing explanation given for the disparate results ing the salience of infant stimuli and attention to it. is the notion that HPA function in depression may Later in the postpartum period (i.e., five to six differ depending on depression subtype (e.g., mel- months postpartum) or in mothers who are stressed ancholic, atypical) or on the history of early adversity and/or ‘at risk’, however, the HPA axis may have a (Gunnar & Fisher, 2006; Heim et al., 2004). negative relation with mothering (Krpan, Coombs, Zinga, Steiner, & Fleming, 2005). In short, the pre- Early adversity and HPA function cise relation between the HPA system and mothering is unclear and likely varies as a function of the age of There is little question that adverse early experiences the mother, the postpartum period being assessed, (or early adversity) can lead to heightened stress the source of the HPA activity (whether through reactivity that in many cases persists into adulthood. stress or not) and the early experiences of the mother Evidence from animal work suggests that early life (Gonzalez et al., 2009; Krpan et al., 2005; Stallings, experiences related to maternal separation and Fleming, Corter, Worthman, & Steiner, 2001). quality of early maternal care modulate the develop- ment of a number of endocrine systems, including the stress system or HPA axis (Champagne & Meaney, Depression and HPA function 2001). Rat moms who were raised without their Activity in the HPA axis has been linked to post- mothers or who were raised by mothers who partum depression, albeit inconsistently. Increased expressed low levels of maternal licking show eleva- cortisol plasma levels have been associated with tions in postpartum glucocorticoids (Burton et al., postpartum depression in some studies (Nierop, 2007; Francis, Diorio, Liu, & Meaney, 1999). Bratsikas, Zimmermann, & Ehlert, 2006), but not in Moreover, maternal separation is associated with others (Groe¨r & Morgan, 2007; Jolley, Elmore, Bar- increased CRF mRNA expression and sensitization nard, & Carr, 2007; Kammerer, Taylor, & Glover, of neurons in hypothalamic and limbic regions, 2006) and have even been associated with positive decreased glucocorticoid receptor density in the affect (e.g., Groe¨r, 2005). As with studies of non- hippocampus and prefrontal cortex, decreased neu- depressed mothers, work suggests that when rogenesis in the hippocampus, as well as altered examining the relationship between cortisol, mater- function of important neurotransmitters (reviewed in nal behavior and mood, factors such as the context Heim et al., 2004). Work on early adversity in pri- for cortisol elevation (e.g., relationship between early mates, where maternal separation or stress para- life stress and cortisol), maternal age and post- digms more closely resemble what happens between partum stage must be considered. More clues as to human moms and infants, has been a bit more com- the relationship between postpartum depression, plex, demonstrating both HPA axis hypoactivity cortisol and early experience may be found in the (Coplan et al., 1996; Rosenblum et al., 2002) and literature on HPA function and non-postpartum hyperactivity (Fahlke et al., 2000; Higley, Suomi, & depression. Abnormal HPA-axis function has been Linnoila, 1991, 1992). Results with human children

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. 376 Jennifer Barrett and Alison S. Fleming and adults map onto some animal HPA axis findings Brown & Moran, 1994; Brown, Cohen, Johnson, & quite nicely. Attachment status has been found to Smailes, 1999; Harkness & Monroe, 2002; Lizardi influence the reactivity of cortisol secretion in infants et al., 1995; McCauley et al., 1997; Moskvina et al., (Gunnar, Brodersen, Nachmias, Buss, & Rigatuso, 2007). Some factors thought to mediate the rela- 1996; Nachmias et al., 1996) and adverse early tionship between early adversity and depression are experience has been shown to relate to adrenal re- gender (women are more vulnerable), the timing of sponse to stress in adolescence (Rao et al., 2008). In the early adversity (earlier childhood is a more vul- adults, in a series of recent studies we found that nerable period), the experience of stress in later life mothers who were either at-risk teens (Krpan et al., and genes (Heim & Nemeroff, 1999; Heim et al., 2005) or clinically depressed (Gonzalez, Steiner, & 2004). Fleming, in preparation) were more likely to show less Generally speaking, the effects of early adversity affectionate and disrupted interactions with their in- on physical and emotional health beginning in early fants and had elevated basal cortisol levels. Teen childhood and continuing throughout development mothers with this endocrine and behavioral profile are believed to be related to plasticity of the central were also more likely to experience inconsistent care nervous system as a function of environment and and to have experienced multiple and changing care- experience. As already noted, stress or trauma givers (Krpan et al., 2005). In addition, adult mothers experienced during development can permanently who were exposed to early adversity (inconsistent care shape brain regions that mediate stress and emo- and/or maltreatment) had higher levels of diurnal tion, including the prefrontal cortex and amygdala cortisol (measured from an area under the curve over (Heim & Nemeroff, 1999; Roceri et al., 2004). At the two consecutive days) and were less sensitive when pre-clinical level, much work in animals has also interacting with their infants (Gonzalez, Steiner, & demonstrated that the repeated stress and associ- Fleming, in preparation). The question remains as to ated physiology can lead to coordinated changes in whether the elevated cortisol levels in high-risk numerous behavioral, autonomic and endocrine mothers are especially related to the postpartum systems that share features with those observed in period or also present through childhood and adult- depression (for reviews see Heim & Nemeroff, 1999; hood. In fact, a number of retrospective studies have Kaufman, Plotsky, Nemeroff, & Charney, 2000; found that outside of the postpartum period, women McEwen, 2003). A prevailing theory about pathways abused as children do show elevated peak adreno- from early adversity to depression involve the notion corticotropic hormone (ACTH) responses to stress and that repeated or severe stress early in life is associ- more prolonged cortisol responses. This effect seems ated with sustained cortisol responses, which is to be strongly related to the woman’s current mood related to brain damage in stress/emotion circuits, state, however, where abused women without cur- which is related to further impairments in stress and rent depression actually demonstrate decreased emotion responses (Heim et al., 2004). Interestingly, cortisol responsiveness and basal cortisol levels and it has also been proposed that not all depression is abused women with depression demonstrate fea- related to early adversity and that early-adversity tures more consistent with hyperactive HPA function precipitated depression may represent a distinct (Heim et al., 2004). depression subtype (Heim et al., 2004). This sug- gestion may underscore the importance of examining early experiences in order to understand the nature Relation of early adversity to depression and etiology of postpartum mood changes. It is difficult to talk about hormones, early adversity and depression without integrating the three. In fact, it is also challenging to find research papers where V. Functional neuroimaging of mothering these topics are not interrelated in some manner. Identifying regions of interest Many of the neurobiological effects of early adversity resemble those of depression and, not surprisingly, To this point we have described a complex set of early adversity is thought to precipitate depression psychological and endocrine factors that are known or contribute to a vulnerability to become depressed. to affect maternal motivation, attitudes, and behav- One of the strongest risk factors for the development ior. What we have not discussed and what follows of depression in adulthood is a history of childhood is an analysis of the growing literature on the neu- adversity, including parenting factors such as roanatomy of mothering, taken from both the animal abuse, insecure attachment, parental rejection and literature and the recent functional magnetic lack of warmth (reviewed in Heim, Plotsky, & resonance imaging (fMRI) literature. The approach Nemeroff, 2004; Karevold et al., 2009; Penza, Heim, we adopt to understanding that neuroanatomy is to & Nemeroff, 2003). In terms of depression, early consider functional neuroanatomy that underlies adversity is associated with earlier onset, greater mothering itself as well as the psychological systems number of episodes, more chronic course of the that affect mothering and that are discussed above. illness and greater suicidality (Bernet & Stein, 1999; Included among these are systems related to Bifulco, Moran, Baines, Bunn, & Stanford, 2002; maternal affect and depression, reward, cognition,

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. Neurobiological perspectives on mothering 377 and early experience. How these functional and excitatory and inhibitory systems. However, most neural systems interface with the maternal circuitry work in the area has focused on the final common is what we now explore. path for the expression of the behavior, the medial The increasingly widespread use of magnetic res- preoptic area of the hypothalamus (MPOA), and its onance imaging (MRI) as a tool to non-invasively downstream projections into the midbrain (ventral investigate the structure and function of the human tegmental area, VTA), hindbrain (periaqueductal brain has opened up the field of human neurosci- gray; PAG) and sensory, limbic, and cortical systems ence to a broad number of topics, including maternal that project into the hypothalamus (see Figure 1). The behavior. Despite this, there are currently fewer than MPOA region of the hypothalamus contains receptors 20 published investigations of the neuroanatomy of for all the hormones involved in the activation of mothering and very few reports of at-risk mom pop- maternal behavior, including receptors for estradiol, ulations. While recent fMRI work (to be detailed later) progesterone, prolactin, oxytocin, vasopressin, and has made a promising start, there is clearly much opioids (see Numan et al., 2006). Neurons projecting work to be done. Despite the relative paucity of into the MPOA region of the hypothalamus are also human mother studies, when looking to understand involved in other behavioral changes, including the maternal brain in human mothers and when changes in mothers’ affect (amygdala, prefrontal formulating our hypotheses/expectations, we are cortex), stimulus salience (amygdala and striatum/ fortunate to be informed by a large body of work on nucleus accumbens), attention (nucleus accumbens the maternal brain in animal species. and medial prefrontal cortex), and memory (nucleus accumbens, medial prefrontal cortex). Some of these Animal studies. The neuroanatomy of animal sites also contain hormone receptors (amygdala, maternal behavior is based predominantly on work medial prefrontal cortex) and may be the sites where with rats, voles, sheep, and primates (summarized in the periparturitional hormones likely act to change Numan et al., 2006). Taken together, these cross- behavior at the time of parturition (Numan et al., species studies indicate a striking similarity in the 2006). neuroanatomy that underlies mothering. Numan has While the rat/animal models of maternal behavior provided us with an exquisite description and anal- are elegant and informative and have identified some ysis of the functional neuroanatomy of maternal key regions in a putative mothering circuit (e.g., behavior in the rat (see Numan & Insel, 2003; Numan hypothalamus, amygdala, prefrontal cortex, nucleus et al., 2006; Numan, 2007). The circuit involves both accumbens), we know that the brains and behaviors

TOUCH cervical (birth) PARIETAL suckling (nursing) CORTEX Neuropeptides perioral (retrieval) Neurotransmitters Hormones Catecholamines MIDBRAIN Hormones NMDA MOTHER-PUP HYPOTHALAMUS BNST MPOA PVN, SON, VMH MATERNAL BEHAVIOUR INTERACTIONS Hormones, Sensory Neuropeptides Inputs

ODOR OLFACTORY AMYGDALA amniotic fluid, BULBS body odors Me/ Ce/ ( NE, DA CORT BASOLAT Opioids Glutamate GABA) CORTICAL AND PREFRONTAL NUCLEUS Catecholamines SUBCORTICAL CORTEX ACCUMBENS NMDA INPUTS

AFFECT REINFORCEMENT OLFACTORY STIMULUS RECOGNITION SALIENCE EXECUTIVE LEARNING FUNCTION NON-MATERNAL BEHAVIOURS

Figure 1 Functional neuroanatomy mediating maternal and related behaviors in mammals. Neuroanatomical structures include olfactory bulbs, amygdala, nucleus accumbens, bed nucleus of the stria terminalis (BNST), medial preoptic area (MPOA), ventromedial hypothalamus (VMH), periventricular nucleus (PVN), supraoptic nucleus (SON), midbrain, and parietal cortex. Relevant neurochemistry includes the catecholamines norepinephrine (NE), and dopamine (DA), the neuropeptides, and the opioids 2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. 378 Jennifer Barrett and Alison S. Fleming of human mothers and mother–infant interactions appropriate. The study of social behavior and related are far more complex than in lower mammals. Thus, constructs (e.g., social cognition, affiliation, love, while animal work provides one important frame- attachment/bonding, empathy) is concerned with work, we must consider other brain systems as well. our perceptions of and interactions with familiar and As already noted, to mother appropriately requires unfamiliar others. The related fMRI studies have the action of multiple behavioral systems in the used a number of different paradigms using social domains of sensation, perception, emotion and stimuli, including reading empathy-eliciting scripts, motivation, reward, executive function, motor output exposure to stimuli of loved ones and exposure to and learning. As a mother–infant dyad are interact- emotional social stimuli (e.g., faces) with and without ing, the mother is receiving sensory input about how contextual factors. Brain regions that are commonly her child looks, sounds and smells, she has feelings identified in these studies include the amygdala, that may change rapidly in the moment and also anterior cingulate cortex (ACC) and dorsal and ven- a baseline level of affect that may not be situa- tral striatum (e.g., Kim et al., 2009a; Kirsch et al., tion dependent, she is moving, touching, kissing, 2005; Platek & Kemp, 2009; Zahn, de Oliveira- speaking, singing, rocking, reacting, she is attending Souza, Bramati, Garrido, & Moll, 2009); these brain and sometimes spreading her attention across a regions are more generally implicated in emotion number of activities or stimuli, she is thinking about perception/regulation, action initiation and stimu- her child and perhaps other things and sometimes lus valence/salience and are also known to contain planning and controlling the nature of play. She is receptors for hormones thought to be important in busy. Of course, as for essentially all experimentally social behavior (i.e., oxytocin, vasopressin; Debiec, manipulated or scientifically observable human 2005; Huber, Veinante, & Stoop, 2005; Keverne & behavior, every region of the brain and all major Curley, 2004; Schorscher-Petcu, Dupre´, & Tribollet, brain systems are active, in some manner. The brain 2009; Smeltzer, Curtis, Aragona, & Wang, 2006). does not have an off switch. Studies of social cognition typically involve response One strategy to reduce this complexity adopted to emotional and non-emotional human faces (e.g., by us (and others), however, is to focus on neural Kirsch et al., 2005; Mitchell, Neil Macrae, & Banaji, systems that are particularly relevant or compelling 2005; Schulz et al., 2009). Kirsch and colleagues in the context of mothering: (a) social behavior; (b) (2005) added the administration of the ‘affiliative’ reward and affect; (c) executive function; and, of peptide, oxytocin, to a social cognition paradigm course, (d) maternal behavior. Studies examining which involved responding to an emotional (angry, maternal behavior will be discussed last, not as a afraid) face-matching task. Intriguingly, the authors reflection of their (in)significance but rather to found that the administration of oxytocin attenuated emphasize the broad overlap between work investi- the expected amygdala response to the negative gating maternal behavior proper and work investi- social stimuli (faces). Amygdala response to emo- gating strongly related psychological constructs. tional or other self-referential faces is complex with After we have examined systems we feel contribute to increased activity in the amygdala associated with and interact with systems mediating maternal negative faces (e.g., fearful faces; Gamer & Bu¨chel, behavior, we can more thoughtfully discuss imaging 2009), positive faces (Derntl et al., 2009; Killgore & studies looking at maternal response to infant Yurgelun-Todd, 2004; Williams, McGlone, Abbott, & stimuli in the context of regions of interest (see Mattingley, 2005) and familiar faces (Platek & Kemp, Figure 2): Hypothalamus/MPOA (maternal behavior, 2009). In contrast, some studies also report a drive states), nucleus accumbens (stimulus sal- decrease in amygdala activity in response to familiar ience, affective valence), amygdala (emotion/affect faces (e.g., Bartels & Zeki, 2004). These discrepant response and regulation, affective valence), anterior findings could reflect the multifaceted role of the cingulate cortex (affect response and regulation, amygdala in the processing of emotional and/or affective valence, stimulus salience, executive func- social stimuli: the amygdala is involved in basic tion, social cognition), orbitofrontal cortex (stimulus stimulus perception, the integration of stimuli with salience, affective valence, executive function), dor- an affective ‘value’ and organizing output to guide solateral prefrontal cortex (affect regulation, execu- affective, autonomic and behavioral responses tive function). To aid the reader in this anatomical (reviewed in Gobbini & Haxby, 2007). The amyg- journey we include a schematic of the neural regions dala’s function in integrating multiple brain systems that will be discussed as well as a table listing the in order to guide behavior is also sensitive to con- different neural structures, their known functions, textual demands (e.g., task requirements, environ- and the relevant references (see Figures 2 and 3). ment), stimulus valence as well as, perhaps, the mood and affect of the individual. Thus, as with animal models, a positive relationship between Social behavior, reward and affect activity in the amygdala and positive mothering Social behavior. As mother–infant interaction is, behavior cannot be assumed and numerous factors broadly speaking, a social behavior, a focus on should be considered when evaluating the role of the mothering within the context of social behavior is amygdala in maternal behavior.

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. Neurobiological perspectives on mothering 379

Figure 2 Brain structures thought to be important in human maternal behaviour. Note that these structures are comprised of classic maternal regions such as the hypothalamus, as well as regions with primary functions important to maternal behaviour. The top panel illustrates the approximate anatomical location of each region; the bottom table contains a coded legend describing each region in terms of location and function

Investigating more abstract concepts related to condition of simply viewing a set of sad faces and an social behavior, a recent study by Kim and col- active condition where participants were instructed leagues (2009a) examined the neural correlates to view the face while attempting to feel/share/ related to compassionate concerns which was understand the person’s suffering. Viewing neutral defined as ‘related to the motivation to approach, and sad faces with a ‘compassionate attitude’, com- help, comfort or otherwise alleviate suffering’ (Kim pared with passive viewing, was related to increased et al., 2009a). During fMRI, male and female par- activity in the medial prefrontal cortex (including ticipants completed a task that involved a passive subgenual ACC), dorsal and ventral striatum as well 2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. 380 Jennifer Barrett and Alison S. Fleming as the insula, thalamus and periaqueductal grey. partners as well as for offspring. In an influential The authors interpreted their results in the context study, Bartels and Zeki (2000) found that viewing of Theory of Mind, which has been associated, pictures of romantic partners compared with friends among other areas, with activity in the medial pre- was related to increases in activity in the striatum, frontal cortex (Saxe, Moran, Scholz, & Gabrieli, ACC, periaqueductal grey, insula and hippocampus 2006; Schulte-Ru¨ther, Markowitsch, Fink, & Piefke, and, interestingly, decreases in activity in regions of 2007; Spreng & Grady, 2009). As previously noted, the prefrontal cortex (Bartels & Zeki, 2000). It is of Theory of Mind, a theory about how we understand note that this pattern of activity was also observed, the intentions and emotions of others, has obvious in a separate study, for own children compared with implications for individual differences in mothering, familiar children (Bartels & Zeki, 2004). The authors where correctly identifying and responding to infant conclude that love relationships are (should be) signals is key. One aspect of Theory of Mind rea- inherently rewarding and must be to ensure the soning is that we understand the internal states of perpetuation of the species. Thus, human attach- others (and hence experience empathy) by simulat- ment may be associated with increased activity ing their experience at the neural level (Blakemore & in reward-related brain regions and, perhaps, Decety, 2001; Rizzolatti & Fabbri-Destro, 2008; decreased activity in brain regions involved in Schulte-Ru¨ther et al., 2007). An extension of this negative affect and/or negative social evaluation. notion is that our ability to understand the mental Indeed, in humans as in animals, there has been world of another is tied to mirror neuron activity, much support for the involvement of dopamine-rich that is, brain response that is related to observation regions of the striatum in feelings of love and of another that mimics the brain response associated attachment (Fisher, Aron, & Brown, 2005; Aron with the (other) occurring thought/emotion. This et al., 2005). idea was first described as a motor phenomenon (see the excellent work of G. Rizzolatti; for a review see Reward and affect. Mothering and nurturance are Rizzolatti & Craighero, 2004), but it has recently also often grouped within basic drive states of sexual been applied to social cognition (e.g., Rizzolatti & desire, hunger/thirst, fear, power-dominance, etc. Fabbri-Destro, 2008; Schulte-Ru¨ther et al., 2007). (Sewards & Sewards, 2003). As already mentioned, Recent work has also linked Theory of Mind to the as infant (and species) survival depends on it, we Default Mode Network, a core set of brain regions expect that most mothers will be motivated to display involved in reflecting on the past, present and future maternal care to some extent and given the right and on the minds of others that also includes the circumstances and history, will find themselves medial prefrontal cortex as a main player (Spreng & more attracted to their babies than repulsed by Grady, 2009). them. That is, with birth, the balance of approach to Relatedly, Zahn and colleagues (2009) were also avoidance will tip in favor of approach and increas- interested in the role of the subgenual ACC in ingly so with additional positive experience. In empathic concern and noted that this region may animal models, maternal behavior has been framed be particularly important when stimuli are self- in the context of approach-avoidance tendencies, relevant and/or reflect individual differences in where the onset of maternal behavior occurs when empathic behavioral response. In this study, a the tendency to avoid pups, find them aversive and/ behavioral index of empathic concern was obtained or view them with indifference is suppressed and the from male and female participants and was later tendency to approach pups and find them rewarding correlated with subgenual ACC response to reading and/or pleasant is enhanced (Fleming & Rosenblatt, scripts designed to elicit feelings of guilt compared 1974a, 1974b; Numan, 2007; Rosenblatt, 1969, with a control condition. The authors reported that 1975a). In the animal model, brain systems that subgenual ACC activity while reading the guilt- initially mediate fear and avoidance include the eliciting scripts was positively associated with an olfactory bulbs and tracts, the medial amygdala, the individual’s level of empathic concern, with higher bed nucleus of the stria terminalis, and the ventro- levels of empathic concern related to greater activ- medial hypothalamus (VMH); prior to late gestation, ity in the subgenual ACC. The ACC is an especially these systems inhibit the MPOA. However, with interesting region as its patterns of connections hormonal priming during late pregnancy and with make it an ideal structure to integrate information parturition, these systems come under inhibition about the internal milieu to guide action, subjective and pup odors which prior to birth were initially feeling or thought. The subgenual ACC in particu- avoided take on a positive valence when the pups are lar, with its strong connections with the amygdala born. It is of note that this priming can also occur in and nucleus accumbens, is also believed to play virgin females, to some extent, by prolonged experi- a more general role in affect regulation and ence with or exposure to pups. Thus, the inhibition motivation. that the MPOA is under in the virgin animal is Finally, much research in the area of social removed when gestational/parturitional hormones behavior has sought to investigate the neural corre- are present or when a virgin experiences sensiti- lates that underlie feelings of love for romantic zation. One can think of this as disinhibiting the

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. Neurobiological perspectives on mothering 381 inhibition that the MPOA is under from the amyg- Anxiety is also an important affective state that is dala, through the VMH. This motivational model of relevant to animal models of mothering and is also mothering involves brain systems important in strongly comorbid with depression. In fact, for the stimulus salience and valence, for example, how most part key cortical and subcortical structures positive pups are and how rewarding they become implicated in the stress response (and also, the del- with experience, or how anxiety-inducing pups are. eterious influence of early adversity) also completely There is certainly evidence at the behavioral level overlap with those involved in mothering (e.g., the for changes in infant-related stimulus salience and prefrontal cortex and amygdala; see Kaufman et al., emotion in mothering in human moms. Compared 2000). Much research examining stress and fear- to mothers, non-mothers find infant odors less related processes, including anticipatory anxiety, pleasing and show a different autonomic and emo- have demonstrated a role for the ACC and the tional (i.e., emotion ratings) response to infant cries amygdala (e.g., Barrett & Armony, 2009; Nitschke (Fleming, Steiner, & Corter, 1997b; Giardino et al., et al., 2009; Sarinopoulos et al., 2009) One model 2008; Stallings et al., 2001). In addition, a mother’s of fear/anxiety learning and extinction posits that own feelings of attachment to her baby increase inhibition of the amygdala by the ACC is important over the course of her pregnancy, a phenomenon in learning when a fear or stress response is appro- that may also reflect an increase in or readiness for priate and when it is not. Some work has also linked ‘approach’ behaviors. However, as we are talking deficits in this system to mechanisms underlying the about motivation, affect and reward, terms which development of maladaptive fear responses as in have been used interchangeably in some studies, post-traumatic stress disorder (reviewed in Rauch, we must emphasize that, in our conceptualization, Shin, & Phelps, 2006), the pathology of which is mothering represents a motivated behavior that has thought to share similar mechanisms with early an impact on and interacts with reward and affect adversity. systems. Given the motivational model of mothering in the Not surprisingly, in humans, the findings of rat, it will be interesting to examine further whether neuroimaging studies focusing on reward and affect or not there exists an infant-related avoidance show considerable overlap with brain regions that tendency in human women and if a similar ACC- are activated in mothers by infant cues; these inhibition of the amygdala may be related to a shift include the prefrontal cortex (ACC, orbitofrontal from avoidance to approach tendencies. An obvious cortex (OFC)), striatum (nucleus accumbens), problem with this suggestion is that human women amygdala and hypothalamus. In a striking example rarely show overt avoidance of infants and in fact the of the overlap between mothering neural circuits opposite is often true for both nulliparous and mul- and those involved in affect (mood), Johansen-Berg tiparous mothers (see Glocker et al., 2009). However, and colleagues (2008) examined the connectivity of differences do exist in terms of emotional and auto- the ventral ACC, the anatomical target of deep brain nomic response to infants in nulliparous and mul- stimulation that has alleviated depression in some tiparous mothers and there are often vast differences patients (Mayberg et al., 2005). The anterior cin- in the nature of responsivity to infants across gulate cortex has been widely implicated in the mothers. It is also important to note that imaging regulation of mood, including major depression, as studies have suggested that human attachment may well as anxiety (Drevets, Savitz, & Trimble, 2008). be associated with increased activity in reward- Using tractography of diffusion imaging data, which related brain regions and, perhaps, decreased allows for the in-vivo mapping of white-matter activity in brain regions involved in negative affect tracts, the authors confirmed in humans what has and/or negative social evaluation (Bartels & Zeki, been established in animals: the anatomical con- 2004). It is possible, then, that a similar motivational nections between the ACC and the hypothalamus, push–pull exists in human mothers, albeit to a more nucleus accumbens, amygdala, OFC, dorsomedial subtle extent. In other words, in a human model the frontal cortex and frontal pole (Johansen-Berg et al., ‘push’ side of things may not be characterized by the 2008). Depression is characterized by impairments unmistakable hostility or avoidance that has been in (or depression of ) other drive states that oper- observed in inexperienced or virgin rats, but by ate on the same neuroendocrinoligical systems as negative affect and anxiety or more subtle signs of maternal behavior (e.g., sexual desire, hunger/ rejection (e.g., looking away from the infant, negative thirst, sleep) and is also characterized by impair- comments about the infant). It is also possible that ment in reward processing (e.g., pleasure of interest functional differences in important regions in such a in people and activities) which is known to be crit- push–pull circuit, for example hyperactivity of the ical in the activation of maternal behavior. Thus, it amygdala and/or lack of inhibition by the ACC, in is not surprisingly that mothering and depression response to the infant, may explain variability in share so much overlap in terms of brain systems, quality of mothering behaviors. nor is it surprising that depression has such a In terms of what is known about the neurocir- powerful impact on the capacity for, or motivation cuitry of reward, a great deal of focus has been on to, mother. the dopamine system and neural structures rich in

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. 382 Jennifer Barrett and Alison S. Fleming dopamine receptors or key structures in dopamine- individuals who are reward sensitive (Carter et al., related brain circuits. Based on animal work, these 2009). regions include dopamine centers in the midbrain It is of note that an equally complicated formula- and the striatum (e.g., nucleus accumbens) and tion has also been proposed for the OFC and reward connected cortical regions (medial prefrontal cor- (see Kringelbach, 2005), which, depending on the tex, OFC; Hollerman, Tremblay, & Schultz, 2000; subregion, is implicated in response to rewards of Wise, 2000). There is now a substantial animal varying complexity (e.g., gain/loss versus taste) as literature showing that the nucleus accumbens is well as positively versus negatively valenced reward important for maternal behavior and that dopa- stimuli. In addition, subregions of the ACC have also mine is released in this site in response to pups or been associated with gain-specific (perigenual ACC), pup cues in new mother rats, experienced mother and gain- and loss-specific (supragenual ACC) rats and virgins induced to be maternal through change in activity in a manner consistent to what simple exposure (i.e., sensitized virgins) (Afonso, has been observed for positive and negative emotions Grella, Chatterjee, & Fleming, 2008; Afonso, King, (Fujiwara et al., 2009). Generally speaking, such Chatterjee, & Fleming, 2009; Champagne et al., distinctions may be important in studies of maternal 2004; Mattson & Morrell, 2005; Olaza´bal, Aber- response using positive and negative infant faces or crombie, Rosenblatt, & Morrell, 2004). Moreover, in mothers with varying sensitivity to reward and there is substantial evidence that the dopamine affect stimuli (related to affective state, stress, early release profile to pups in mothers depends on the experience, etc.). Related to this, recent fMRI work action of the parturitional hormones and is differ- has demonstrated that childhood adversity influ- ent from the profile of dopamine release to food in ences the neural (basal ganglia) and behavioral these mothers (Afonso et al., 2009). It is different (emotion rating) response to monetary reward in as well from the profile shown by sensitized virgins, adults, where individuals exposed to childhood but not the profile shown by hormonally primed adversity demonstrated attenuated brain response virgins (Afonso et al., 2008, 2009). to reward cues in the globus pallidus and rated re- Human work has for the most part confirmed the ward cues less positively (Dillon et al., 2009). This (non-maternal) animal findings in imaging studies work nicely complements animal studies that have examining the rewarding nature of food (Goldstone demonstrated a relationship between adverse rear- et al., 2009; Grabenhorst, Rolls, Parris, & d’Souza, ing experiences related to maternal separation and 2009), social acceptance (Davey, Allen, Harrison, altered responses to both rewarding and aversive Dwyer, & Yu¨cel, 2009), money (Fujiwara, Tobler, stimuli (Nelson et al., 2009; Pryce, Dettling, Spen- Taira, Iijima, & Tsutsui, 2009; Simon et al., 2010) gler, Spaete, & Feldon, 2004). and, appropriately, infantile physical features Not surprising, numerous fMRI studies have also (Glocker et al., 2009). While this recent work has demonstrated depression-related alterations in neu- confirmed posited reward structures and circuits, it ral and behavioral response to emotional and/or has also revealed more subtle features and func- rewarding stimuli (Dannlowski et al., 2007, 2008; tionality of specific regions in this circuit. For Dichter, Felder, & Smoski, 2009; Fu et al., 2008; Gray example, Carter, Macinnes, Huettel, and Adcock et al., 2006; Raes, Hermans, & Williams, 2006; (2009) examined the degree to which the striatum Smoski et al., 2009). People with depression exhibit encodes both the motivational and affective rele- negative biases in processing facial expressions, vance of anticipated outcomes. To do this, men and altered activity in emotion-related brain regions in women completed a reward anticipation task during response to emotion faces (e.g., greater amygdala fMRI where monetary gains and loss outcomes response to emotional faces) as well as blunted could be either self- or charity-directed. They found response to positive or rewarding stimuli (Dannlowski that activity in the nucleus accumbens and ventral et al., 2007, 2008; Dichter et al., 2009; Fu et al., 2008; tegmental area was increased in anticipation of both Gray et al., 2006; Raes et al., 2006; Smoski et al., gain and loss trials, supporting the notion that 2009). In a recent study investigating the reward these regions are sensitive to the motivational rele- system, Pizzagalli and co-authors (2009) examined vance (salience), not the valence, of an event. activity in the striatum in patients with depression Evidence regarding the specificity of response in the while they completed a monetary reward task. As nucleus accumbens to stimulus salience versus expected, compared to control participants, partici- valence is unclear, however, as it has also been pants with depression showed blunted response to demonstrated that accumbens activity correlates monetary gain in the nucleus accumbens. Interest- with both gains and losses only in conditions of ingly, depression severity was also associated with uncertainty (Cooper & Knutson, 2008). An attempt reduced volume in the caudate nucleus. has been made to reconcile these findings by noting that while the nucleus accumbens may be primarily Executive function sensitive to stimulus salience (e.g., reward-based or not reward-based), its response may be influenced We know that optimal mothering requires that by affective valence (e.g., gain or loss), especially in the mother pays attention to her infant, that

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. Neurobiological perspectives on mothering 383 she responds appropriately to her infant and anterior cingulate cortex has been widely linked to that she responds contingently to the infant’s cries tasks involving attention and action monitoring, and behavior. As indicated earlier, she must be especially in the context of error or conflict (Barch, flexible and be able to focus when appropriate and Braver, Sabb, & Noll, 2000; Brown & Braver, 2005, switch attention when the situation changes. And 2008; Modirrousta & Fellows, 2008; van Veen & finally, she must retain information about the infant Carter, 2002; Wang, Ulbert, Schomer, Marinkovic, & and adequately manipulate the environment to Halgren, 2005) and the orbitofrontal cortex has been optimize it for the infant. In light of these cognitive linked to impulse control (Horn, Dolan, Elliott, demands, the activity of brain regions mediating Deakin, & Woodruff, 2003; Matsuo et al., 2009; planning, attention, flexibility and working memory reviewed in Fineberg et al., 2010; Winstanley, are essential for adequate mothering and thus con- Theobald, Cardinal, & Robbins, 2004). While the stitute the focus of a number of maternal studies. relationship between prefrontal cortex function and Given its role in attention, cognitive flexibility and maternal behavior has not been widely investigated, working memory, the dorsolateral prefrontal cortex there is considerable evidence that both early (DLPFC) appears an ideal structure to help identify adversity and non-postpartum depression have an and attend to infant cues in a complex environment, impact on the structure and/or function of the efficiently switch-tasks and attention and maintain/ prefrontal cortex (animal studies: Banasr et al., manipulate information for on-line decision-making 2007; Roceri et al., 2004; Roth, Lubin, Funk, & and action planning. Over the past decade or so, Sweatt, 2009; human studies: Clark, Chamberlain, there has been much debate regarding the function & Sahakian, 2009; Hercher, Turecki, & Mechawar, of the DLPFC and whether this region is organized in 2009; Koenigs & Grafman, 2009; Rogers et al., 2004; a ‘domain-specific’ (i.e., visuo-spatial versus verbal) Savitz & Drevets, 2009; Siegle, Thompson, Carter, or ‘process-specific’ (i.e., monitoring versus retrieval) Steinhauer, & Thase, 2007) and in some cases manner. Supporting the latter, the detailed work of structure and function have been linked. For exam- Michael Petrides has contributed much to our ple, in rats, early experience-mediated deficits in understanding of the function of the DLPFC (specif- executive function-processes such as prepulse inhi- ically, the mid-dorsolateral prefrontal cortex, corre- bition (PPI) and set-shifting (Burton et al., 2006; sponding to areas 46 and 9/46), which he sees as: Garner et al., 2007; Lovic & Fleming, 2004) are a system for the conscious active control of plan- accompanied by structural changes in the pre-fron- ned behavior and cognition. Such a system must have tal cortex (PFC) (Burton et al., 2007; Chatterjee et al., the capacity to hold abstract coded representations of 2007). Further work is needed, however, to confirm events that are expected to occur, so as to mark their the predicted importance of prefrontal function in occurrence or non-occurrence (i.e. monitor their rela- maternal behavior both outside of and in the context tive status in relation to each other and the intended of important individual differences such as early plan). Furthermore, such a system would be involved in experience and mood. the manipulation of these cognitive representations (i.e. planning) since such manipulation would require con- stant monitoring of the relative status of intended acts Maternal behavior or events (Petrides, 2005). Work on the neural bases of maternal behavior in Although executive functions rely on interactions humans is derived primarily from fewer than 20 fMRI between many brain regions (for reviews see Duncan studies where mothers, non-mothers, and some- & Owen, 2000; Leh, Petrides, & Strafella, 2009), times fathers are presented with either pictures of numerous neuroimaging studies in humans have infants (Barrett et al., 2009; Bartels & Zeki, 2004; highlighted an important role for the DLPFC in tasks Leibenluft, Gobbini, Harrison, & Haxby, 2004; that involve task- or set-shifting (e.g., Monchi, Nitschke et al., 2004; Strathearn, Fonagy, Amico, & Petrides, Petre, Worsley, & Dagher, 2001; Nagahama Montague, 2009; Strathearn, Li, Fonagy, & Monta- et al., 2001; Ravizza & Carter, 2008), auditory and gue, 2008), recorded infant cries (Lorberbaum et al., spatial attention (Krumbholz, Nobis, Weatheritt, & 2002; Seiftriz, 2003; Swain et al., 2008), or video- Fink, 2009), attention related to perceptual selection tapes of infants (Noriuchi, Kikuchi, & Senoo, 2008; (Nee & Jonides, 2009), working memory involved in Ranote et al., 2004). For thorough and insightful the Iowa Gambling Task (Li, Lu, D’Argembeau, Ng, & reviews of the literature see recent papers by James Bechara, 2009), solving anagrams with varying E. Swain and colleagues (e.g., Swain, 2008; Swain, degrees of abstraction (Christoff, Keramatian, Gor- Lorberbaum, Kose, & Strathearn, 2007). By exam- don, Smith, & Ma¨dler, 2009), memorial selection ining these studies (many are reviewed in Swain & (Nee & Jonides, 2009) and active maintenance of Lorberbaum, 2008 and Swain et al., 2007), we can information for ongoing task performance (Cole & get an idea about the specificity of our posited Schneider, 2007). maternal systems of interest to infant type (own, Other regions of the frontal lobes that are relevant other), stimulus type (infant vs. non-infant) as well to maternal behavior also have prominent roles in as stimulus familiarity, modality (auditory, visual, executive functions: the ACC and OFC. The dorsal both) and even subject (mother, non-mother). For the

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. 384 Jennifer Barrett and Alison S. Fleming most part, the findings of fMRI work on the maternal and the hormones of parturition are not as influen- brain demonstrate that irrespective of stimulus tial on the brain and behavior, the function and/or modality (e.g., visual or auditory) or stimulus famil- response of maternal neural systems may be quite iarity (see Leibenluft et al., 2004), many of our different. Given this possibility, examining maternal maternal regions of interest (e.g., hypothalamus, response to children with a broad range of ages may nucleus accumbens, amygdala, anterior cingulate further add to unwanted ‘noise’ in the observed cortex, orbitofrontal cortex, dorsolateral prefrontal maternal system. cortex) are implicated in response to a mother’s own Discrepant findings across studies may also be infant compared to another infant or non-infant attributed to differences in the experimental para- control. In studies comparing parents to non- digm, or specifically, in the comparison used to best parents, neural response to infants in maternal capture maternal response. Work on the human regions have also been found to be specific to maternal brain rests on the acceptance of the notion parental status (e.g., Seifritz et al. 2003, but see that we are able to elicit a response in the fMRI Glocker et al., 2009 for a demonstration of how scanner that is relevant to and reflective of some- infant schema are rewarding for nulliparous thing as complex as mothering. Capturing maternal women). That is, as we would expect, one’s own behavior or maternal responsiveness in an fMRI infant is a special stimulus that engages brain scanner is a considerable challenge, however, and as regions that we know are important in maternal previously noted, choice of a particular paradigm behavior and related systems (e.g., executive func- has many implications for how we interpret the tion, reward, affect). As the role of these systems in results and the likelihood that a particular system or maternal behavior has not yet been addressed in a structure will be engaged. The choice of approach is detailed manner, we cannot yet determine the func- at the discretion of the investigators and thus they tional significance of identified maternal regions in tend to vary considerably. For example, some studies the context of the many domains involved in moth- have used an unfamiliar infant compared to a con- ering. Indeed, the current state of literature on the trol stimulus (e.g., computer-generated noise, house) human maternal brain holds many more questions to capture maternal response; most studies use than answers and we find that there are large dif- mothers’ own children and contrast responses to ferences across studies in what and how ‘maternal’ them with that of unfamiliar or familiar children. regions are implicated. In what follows we briefly These approaches all have different strengths and discuss a few key areas that could influence brain weaknesses in regard to ecological validity as well as imaging results. Also, we talk about how identifying what can be inferred from the imaging results. One and understanding important individual differences may argue that using a standard infant stimulus may help move this area of research forward. (not mother’s own infant) in order to capture mater- Although no imaging paradigm will give us a nal response may not be as effective in engaging the spotlight on a maternal circuit, different patterns of maternal circuit, the function of which we assume is results observed across maternal imaging studies at least in part specialized for or uniquely connected can be attributed to factors such as the age of chil- to the mother’s own offspring. That being said, we dren studied (newborn to early childhood) and the must concede that it is also possible (and likely) that comparison used to best capture the maternal brain certain regions within a maternal circuit will respond (own baby compared to control stimuli, familiar to any infant stimuli, and perhaps even find such infant or unfamiliar infant). In regard to the former, stimuli rewarding compared to a non-infant stimu- the early postpartum period is a unique time, both lus, especially for a mother in comparison to a non- hormonally and behaviorally; the mother is learning mother (see Glocker et al., 2009). Another major much about her infant at this time, attachment question in terms of studying the maternal circuit is patterns are forming, the maternal system may be in regard to the most optimal comparison stimulus especially primed for action and one would predict for a mother’s own infant. That is, to isolate brain that this would be a time of great activity for brain activity associated with a maternal response to one’s systems important in maternal behavior. In fact, a own infant while controlling for confounding factors main function of the maternal system in the early or factors of no interest (i.e., the ‘infantness’ of the postpartum period is ensuring that the mother will face, familiarity of the stimulus, emotionality of the respond to her infant, will find her infant pleasing stimulus). Strategies that have been used include a and is attracted to her infant, is attentive, sensitive, non-infant stimulus (e.g., an object, artificially cre- etc. There are great differences, of course, across ated sound), an unfamiliar baby or a familiar baby. mothers in terms of the quality and quantity of these Using an unfamiliar infant as a control stimulus maternal behaviors and we predict that these dif- takes care of many confounds, with the exception of ferences are related to differences in the function of familiarity (of own baby) and novelty (of unfamiliar the maternal system. One may further predict that baby) which could bias results in manner that is not later on and in the postpartum period and through- necessarily reflective of maternal response (e.g., out the life of the child, once the nature of mother– Leibenluft et al., 2004). Furthermore, when com- child relationship is more habitual and consolidated paring brain response to own and familiar infants to 2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. Neurobiological perspectives on mothering 385 which there are likely varying degrees of emotion and patterns in the imaging literature related to meth- maternal feelings, there is also the risk that differ- odology or maternal experience that predict a role for ences in brain regions of interest (especially small some maternal regions and not others. In mothers, subcortical regions) will be too subtle to be detected. factors have been found to mediate maternal Thus, the main message we wish to convey is that response to infant stimuli, however, including type of the ‘perfect’ paradigm to identify a maternal circuit delivery (Swain et al., 2008) and adult attachment does not exist and that we have at our disposal a status (Strathearn et al., 2009). Swain and co- number of options with strengths and weaknesses authors (2008) examined the effect of type of delivery that will influence the relevance (in terms of regions on the maternal response to own infant cries com- of interest) of brain imaging findings to something as pared to an unfamiliar infant cry and temporally complex as mothering. matched white-noise. A main rationale behind Another factor that may impact the nature of examining type of delivery was that the experience of maternal responsivity in brain imaging studies is childbirth by vaginal delivery (VD) compared with maternal experience and parity. Patterns of brain cesarean section delivery (CSD) uniquely involves response as a function of mothers’ parity or maternal the release of oxytocin (Swain et al., 2008). Using a experience have not yet been adequately investigated similar design to that of Lorberbaum et al. (2002), six in humans. However, based on the animal work we CSD and six VD mothers completed a block-design can make a number of predictions. In the rat work fMRI task 2–4 weeks postpartum. A region-of-inter- we know that certain brain regions are more active est analysis revealed that in response to their own (established through expression of the c-fos gene) in infant cry compared to an unfamiliar infant cry, experienced than inexperienced animals when pre- VD mothers showed a greater response in a number sented with pup stimulation. These areas include of regions including the hypothalamus, striatum and sensory areas important for mothering in the rat and amygdala. It was noted that the results for the here the parietal cortex which mediates touch stim- hypothalamus in particular may highlight its role in ulation is implicated. Other areas that show greater mothering and mothering behaviors such as activation in multiparous or experienced animals breastfeeding as well as attachment (Swain et al., over primiparous animals include the MPOA (the 2008). In a recent study, Strathearn and colleagues so-called final common pathway) and the anterior (2009) examined brain response to happy, sad and cingulate region (Fleming & Korsmit, 1996). neutral own and unfamiliar infant faces in mothers Recent animal work also shows that the parity whose attachment status was defined as secure or differences that are seen in brain activation patterns insecure/dismissing. A number of different mea- relate more to the presence and absence of mother- sures, including questionnaire data, videotaped ing experience than to variations in mothers’ hor- mother–infant interaction and blood collection for monal profile, although hormones may well enhance oxytocin sampling, were collected over an approxi- or intensify brain responses to pups in maternal mately one-and-a-half year period. The fMRI session animals. For instance, virgin rats who become was conducted at seven months postpartum. Their maternal through sensitization to pup stimulation, main findings were that compared to insecure/dis- through simple exposure, show the same pattern of missing mothers, securely attached mothers dem- brain activation as do primiparous mothers who onstrated significantly more activity in the lateral become maternal through hormonal mechanisms. prefrontal cortex, medial prefrontal cortex and the The pattern of effects is the same in the two cases, hypothalamus/pituitary region in response to their although the hormonally primed primiparous animal own infant faces and greater response in the ventral shows a more ‘robust’ activation pattern than does striatum, orbitofrontal cortex and medial prefrontal the sensitized virgin animal when presented with cortex in response to own happy infant faces. Inter- pups, suggesting a role for the hormonal activation estingly, the authors also found that insecurely/ of behavior (Numan et al., 2006). And what of the dismissing mothers had greater activity in brain human analogy of mothers who are adopting regions associated with negative affect (e.g., the infants? Although, among humans, comparisons in insula) in response to their own infant’s sad faces. brain response patterns between adoptive and non- These two studies have nicely demonstrated how adoptive parents to infant cues have not been engagement of a maternal system by infant stimuli explored, based on the animal work one might is not universal and is in fact influenced by mater- predict a pattern of activation in adoptive mothers nal factors and individual differences relevant to that is similar to the pattern seen in new mothers. mothering. Furthermore, it is likely that individual Anecdotal evidence suggests that adoptive parents differences that are related to circuits that overlap and are certainly as motivated as biological parents with or interact with maternal circuits, such as and experience strong positive affect and feelings of depression and early experience, may be particu- attachment. This is certainly true of alloparents larly influential in how a mother’s brain responds to (see Hrdy, 2009). her infant. The above discussions are for the most part Although published reports are few at this time, speculative as there are at this time no discernible important individual differences in maternal brain

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. 386 Jennifer Barrett and Alison S. Fleming structure and function related to early experience inhibition or facilitation of brain process or func- and mood/affect are currently being investigated tion. Based on what is known about the function of (e.g., Barrett et al., 2009; Kim et al., 2009b). A recent the ACC, we may speculate that increased response study by Kim and colleagues examined the rela- in this region is associated with self- or other-ref- tionship between retrospective reports of early erential thoughts or evaluation, with the simulation maternal care and (1) mother’s brain structure (gray of experience, with affective response and/or arou- matter volume) and (2) maternal brain response to sal. Note that all of these processes or states may infant cries. In the group of mothers, perceived vary along an affect dimension, having positive, higher maternal care in childhood was associated neutral or negative valence. While we may predict with higher gray matter volume in frontal, orbital that poorer quality of early parental care would be and temporal cortices (Brodmann areas 10, 47 and related to the negatively valenced side of things in 22/20, respectively) as well as greater brain terms of maternal responsiveness, this suggestion response to infant cries in regions including the warrants further investigation. DLPFC and temporal cortex. Reports of low maternal care were also related to increased brain response in the hippocampus. The authors suggest that their VI. Final integration findings reflect early experience-mediated differ- ences in the ability to understand the infants’ mental The nature of the mother–infant dance is complex and physical states (key to sensitive mothering) as and relies on the integrity and function of numerous well as affective reactivity to cry stimuli. physiological and behavioral systems in the do- In our own imaging work, we are examining mains of sensation, perception, affect, reward, mothers early in the postpartum period (three executive function, motor output and learning. To months postpartum) and collecting indices of engage in mothering behavior, mothers have to be behavioral (coded videotaped mother–infant inter- sensitive to infant cues, select those cues for pro- action), affective (depression and anxiety scales, cessing and thus such cues must be attractive and ratings of affective response to infant faces), attitu- salient for the mother. Furthermore, mothers must dinal (infant-related), hormonal (cortisol) and neural be emotionally prepared and positively motivated to (BOLD response to positive and negative own and engage socially with the infant, they must selectively unfamiliar infant faces) responsiveness as well as attend to the infant in the context of competing retrospective accounts of early experience in family stimuli, and finally, they must be restrained and of origin (parental bonding, childhood abuse). consistent in their responsiveness. The nature of the Although our work is ongoing, one of our analysis mother–infant dance relies on the integrity of mul- strategies has been to extract brain data (i.e., beta tiple physiological and behavioral systems and also weights derived by the general linear model) from a on two maternal experiential factors, one proximal region of interest (the anterior cingulate cortex) and and one distal, that have a great impact on how a relate activity in this region to a maternal factor mother mothers: postpartum depression and early (retrospective reports of early experiences in family experiences. When a mother is at risk to engage in of origin, see Barrett et al., 2009). By linking brain less than optimal mothering, such as when she is structures or circuits to specific maternal factors depressed or has experienced adversity in child- and examining individual differences, we feel this hood, the function of many or all of maternal and approach will be helpful in understanding the related systems may be affected. In this chapter, we neurobiology underlying the many components that have detailed how mothering is intimately inter- comprise and influence how mothers mother. To twined with affect and the deleterious effects of early this aim, our preliminary work in 12 mothers sug- experience (e.g., stress systems). The interrelation gested that maternal retrospective reports of quality between maternal, affect and stress systems mani- of early experiences related to parental bonding and fests in many ways. At the behavioral level, maternal abuse are significantly related to maternal anterior affect and quality of early experiences are related to cingulate cortex response to own baby (compared less than optimal attachment relationships, mater- to unfamiliar baby), where greater magnitude of nal attitudes and mother–infant communication. At anterior cingulate cortex response to one’s own the hormonal level, the neurobiology of these sys- baby is related to reports of poorer parental bonding tems is strikingly similar. Furthermore, it is likely (maternal and paternal care) and more emotional that the similarity and cross-talk between maternal abuse (Barrett et al., 2009). It is difficult at this affect and stress systems also allows for mood dis- time to know what increased ACC response to one’s turbance and early adverse experiences to have a own infant means about the mother and her significant impact on the quality of mothering and responsivity or maternal behavior in the context of the motivation to mother. quality of early experience. In addition, while we The main purpose of this discussion was to may infer that increases in BOLD signal reflect extend what is known about the neurobiology of increases in population activity in a given brain mothering to functional neuroimaging. Instead of region, we are not sure if the activity represents the focusing on strictly ‘maternal’ studies, we instead 2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. Neurobiological perspectives on mothering 387

Figure 3 Schematic of a putative human maternal circuit. Brain regions known or believed to be important in human maternal behavior serve a number of important functions. There are (relatively) gross subcortical functions related to reward, affect and motivation that serve to ensure that the baby is a strongly positively reinforcing stimulus that the mother is driven to care for, to be responsive to and show affection to. There are also higher-order cortical functions that regulate maternal behavior in the context of the (internal and external) environment and in light of over-arching goals, standards, mores and intentions. Note that some regions, namely the anterior cingulate cortex and orbitofrontal cortex, operate at the interface of the two main systems

used a systems approach to survey important without reference to genetics and their impact on findings relevant to brain systems integral to and/ the variety of ways the environment is expressed in or strongly related to the mothering experience: (a) mothers’ behavior (Bakermans-Kranenburg & Van social behavior; (b) reward and affect; (c) executive IJzendoorn, 2008). Some mothers are attentive, function; and (d) maternal behavior. In human others not; some mothers are affectively attached, neuroscience, without ablation techniques, micro- others not. Some mothers are predominantly talk- dialysis or electrical stimulation, direct brain– ers; others are quieter but interact through stroking behavior relationships are difficult to nail down and and touching; others avoid or evade. And mothers we must often extrapolate and speculate. We have differ enormously from one another in their tem- seen that this is especially challenging when peraments, their hormones, and their brains. These examining something as complex as human individual differences cannot be entirely explained maternal behavior. However, by examining each of by knowing the early environment in which mothers our systems of interest in the context of maternal were reared, or by knowing the present social con- behavior, it becomes easier to envision the function text or affective state in which the mother finds of each region in our putative maternal circuit (see herself. Figure 3). Furthermore, we find that there are The notions of psychological resilience in face of many commonalities in terms of the brain regions adversity and of susceptibility to positive effects of identified across these systems and, as we would ‘good’ early experience bring with them questions expect, all are sensitive to the influence of, or about the mother’s genotype and how polymor- function differently in the context of, depression phisms in specific genes and gene sets interact with and adverse early experience. So, in addition to the early and present experience to help create the vast behavioral and hormonal findings, we have identi- differences in temperament, affect, reward suscep- fied another mechanism by which individual dif- tibility, sensitivity and executive function capabili- ferences can impact the quality of mothering. Key ties necessary for mothers to mother effectively. This to progress in mapping the maternal brain will be next decade, when we use what we have learned discovering and accounting for important individ- about the genome to better understand behavior and ual differences that influence brain regions impor- brain, should be a very exciting time for the study of tant in mothering and determining whether or not the psychobiology of mothering. some circuits/regions are influenced more than others and how, in other words, to identify specific structure–function relationships in the context of Correspondence to maternal behavior. As often observed in studies of behavior and hormones, all mothers are not cre- Jennifer Barrett, Department of Psychology, Uni- ated equal and brain differences should, in part, versity of Toronto at Mississauga, South Building, reflect this. Room 2037B, 3359 Mississauga Rd. N., Mississau- To conclude then, although we have learned a ga, ON N5L 1C6, Canada; Tel: 905-569-4292; Email: great deal about the psychobiology of mothering [email protected]; or Alison S. Flem- over the past 20 years, there is still much to learn ing, Department of Psychology, University of Toronto and much to revisit, given new knowledge. No at Mississauga; Tel: 905-828-3961; Email: alison. discussion on individual differences is complete fl[email protected]

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. 388 Jennifer Barrett and Alison S. Fleming

Key points

• Mothering behavior and feelings are a product of a wide variety of behavioral and physiological propen- sities and underlying systems. • Quality of mothering relies on maternal factors, one proximal and one distal, that have an impact on how a mother mothers: postpartum depression and adverse (stressful) early experiences. • The brain mechanisms that regulate mothering include modules or systems within the hypothalamus, the limbic system and the cortex; these systems are interrelated and are sensitive to the influence of, or function differently in the context of, variations in maternal mood and early experiences in family of origin. • In future work, the study of functional, molecular and gene expression changes in the brain that occur under different environmental contexts and/or that vary with genotype may help us to understand why mothers differ from one another in the extent to which their mothering and postpartum mood state is affected by adverse early experiences and stresses in their lives.

Banasr, M., Valentine, G.W., Li, X., Gourley, S.L., Taylor, References J.R., & Duman, R.S. (2007). Chronic unpredictable Abdulrazzaq, Y.M., Al Kendi, A., & Nagelkerke, N. (2009). stress decreases cell proliferation in the cerebral cortex Soothing methods used to calm a baby in an Arab of the adult rat. Biological Psychiatry, 62, 496–504. country. Acta Paediatrica, 98, 392–396. Banki, C.M., Bissette, G., Arato, M., O’Connor, L., & Afonso, V.M., Grella, S.L., Chatterjee, D., & Fleming, A.S. Nemeroff, C.B. (1987). CSF corticotropin-releasing (2008). Previous maternal experience affects accumbal factor-like immunoreactivity in depression and schizo- dopaminergic responses to pup-stimuli. Brain Research, phrenia. The American Journal of Psychiatry, 144, 873– 1198, 115–123. 877. Afonso, V.M., King, S., Chatterjee, D., & Fleming, A.S. Banki, C.M., Karmacsi, L., Bissette, G., & Nemeroff, C.B. (2009). Hormones that increase maternal responsive- (1992). Cerebrospinal fluid neuropeptides in mood dis- ness affect accumbal dopaminergic responses to pup- order and dementia. Journal of Affective Disorders, 25, and food-stimuli in the female rat. Hormones and 39–45. Behavior, 56, 11–23. Barch, D.M., Braver, T.S., Sabb, F.W., & Noll, D.C. Afonso, V.M., Sison, M., Lovic, V., & Fleming, A.S. (2000). Anterior cingulate and the monitoring of (2007). Medial prefrontal cortex lesions in the female response conflict: Evidence from an fMRI study of overt rat affect sexual and maternal behavior and their verb generation. Journal of Cognitive Neuroscience, 12, sequential organization. Behavioral Neuroscience, 121, 298–309. 515–526. Barrett, J., & Armony, J.L. (2009). Influence of trait anxiety Ainsworth, M., Bell, S., & Stayton, D. (1974). Infant– on brain activity during the acquisition and extinction of mother attachment and social development: Socializa- aversive conditioning. Psychological Medicine, 39, 255– tion of reciprocal responsiveness to signals. In M. 265. Richards (Ed.), The integration of a child into a social Barrett, J., Hall, G., Wonch, K., Ali, N., Gonzalez, A., world (pp. 99–135). Cambridge: Cambridge University Steiner, M., et al. (2009). Quality of early parental care Press. reported by mothers is related to anterior cingulate Anisfeld, E., Casper, V., Nozyce, M., & Cunningham, N. response to own infant stimuli: Preliminary findings. (1990). Does infant carrying promote attachment? An NeuroImage, 47, S181–S181. experimental study of the effects of increased physical Bartels, A., & Zeki, S. (2000). The neural basis of romantic contact on the development of attachment. Child Devel- love. Neuroreport, 11, 3829–3834. opment, 61, 1617–1627. Bartels, A., & Zeki, S. (2004). The neural correlates of Arnott, B., & Meins, E. (2007). Links among antenatal maternal and romantic love. NeuroImage, 21, 1155– attachment representations, postnatal mind-minded- 1166. ness, and infant attachment security: A preliminary Beebe, B., Badalamenti, A., Jaffe, J., Feldstein, S., Mar- study of mothers and fathers. Bulletin of the Menninger quette, L., Helbraun, E., et al. (2008). Distressed moth- Clinic, 71, 132–149. ers and their infants use a less efficient timing Aron, A., Fisher, H., Mashek, D.J., Strong, G., Li, H., & mechanism in creating expectancies of each other’s Brown, L.L. (2005). Reward, motivation, and emotion looking patterns. Journal of Psycholinguistic Research, systems associated with early-stage intense romantic 37, 293–307. love. Journal of Neurophysiology, 94, 327–337. Belsky, J. (1984). The determinants of parenting: A process Atkinson, L., Leung, E., Goldberg, S., Benoit, D., Poulton, model. Child Development, 55, 83–96. L., Myhal, N., et al. (2009). Attachment and selective Benoit, D., & Parker, K.C. (1994). Stability and transmis- attention: Disorganization and emotional Stroop reac- sion of attachment across three generations. Child tion time. Development and Psychopathology, 21, 99– Development, 65, 1444–1456. 126. Bergman, K., Sarkar, P., Glover, V., & O’Connor, T.G. (in Bakermans-Kranenburg, M.J., & van IJzendoorn, M.H. press). Maternal prenatal cortisol and infant cognitive (2008). Oxytocin receptor (OXTR) and serotonin trans- development: Moderation by infant–mother attachment. porter (5-HTT) genes associated with observed parent- Biological Psychiatry. ing. Social Cognitive and Affective Neuroscience, 3, Bernet, C.Z., & Stein, M.B. (1999). Relationship of child- 128–134. hood maltreatment to the onset and course of major

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. Neurobiological perspectives on mothering 389

depression in adulthood. Depression and Anxiety, 9, Sprague–Dawley rats. Behavioral Neuroscience, 120, 169–174. 665–675. Bernstein, I.H., Rush, A.J., Yonkers, K., Carmody, T.J., Burton, C.L., Chatterjee, D., Chatterjee-Chakraborty, M., Woo, A., McConnell, K., et al. (2008). Symptom features Lovic, V., Grella, S.L., Steiner, M., et al. (2007). Prenatal of postpartum depression: Are they distinct? Depression restraint stress and motherless rearing disrupts expres- and Anxiety, 25, 20–26. sion of plasticity markers and stress-induced corticoste- Bifulco, A., Kwon, J., Jacobs, C., Moran, P. M., Bunn, A., & rone release in adult female Sprague–Dawley rats. Brain Beer, N. (2006). Adult attachment style as mediator Research, 1158, 28–38. between childhood neglect/abuse and adult depression Carmichael, S.T., & Price, J.L. (1994). Architectonic sub- and anxiety. Social Psychiatry and Psychiatric Epidemi- division of the orbital and medial prefrontal cortex in the ology, 41, 796–805. macaque monkey. Journal of Comparative Neurology, Bifulco, A., Moran, P.M., Baines, R., Bunn, A., & Stanford, 346, 366–402. K. (2002). Exploring psychological abuse in childhood: II. Carpenter, L.L., Tyrka, A.R., McDougle, C.J., Malison, Association with other abuse and adult clinical depres- R.T., Owens, M.J., Nemeroff, C.B., et al. (2004). Cere- sion. Bulletin of the Menninger Clinic, 66, 241–258. brospinal fluid corticotropin-releasing factor and per- Blakemore, S.J., & Decety, J. (2001). From the perception ceived early-life stress in depressed patients and healthy of action to the understanding of intention. Nature control subjects. Neuropsychopharmacology, 29, 777– Reviews. Neuroscience, 2, 561–567. 784. Bornstein, M.H., Tamis-Lemonda, C.S., Hahn, C., & Carter, R.M., Macinnes, J.J., Huettel, S.A., & Adcock, R.A. Haynes, O.M. (2008). Maternal responsiveness to young (2009). Activation in the VTA and nucleus accumbens children at three ages: Longitudinal analysis of a mul- increases in anticipation of both gains and losses. tidimensional, modular, and specific parenting con- Frontiers in Behavioral Neuroscience, 3, 21. struct. Developmental Psychology, 44, 867–874. Champagne, F., & Meaney, M.J. (2001). Like mother, like Bos, K.J., Fox, N., Zeanah, C.H., & Nelson, C.A. III (2009). daughter: Evidence for non-genomic transmission of Effects of early psychosocial deprivation on the develop- parental behavior and stress responsivity. Progress in ment of memory and executive function. Frontiers in Brain Research, 133, 287–302. Behavioral Neuroscience, 3, 16. Champagne, F.A., Chretien, P., Stevenson, C.W., Zhang, Boyce, P.M., & Todd, A.L. (1992). Increased risk of post- T.Y., Gratton, A., & Meaney, M.J. (2004). Variations in natal depression after emergency caesarean section. The nucleus accumbens dopamine associated with individ- Medical Journal of Australia, 157, 172–174. ual differences in maternal behavior in the rat. Journal of Brand, S.R., & Brennan, P.A. (2009). Impact of antenatal Neuroscience, 24, 4113–4123. and postpartum maternal mental illness: How are the Champagne, F.A., Francis, D.D., Mar, A., & Meaney, M.J. children? Clinical Obstetrics and Gynecology, 52, 441– (2003). Variations in maternal care in the rat as a 455. mediating influence for the effects of environment on Brazelton, T.B. (1977). Implications of infant development development. Physiology and Behavior, 79, 359–371. among Mayan Indians of Mexico. In P.H. Leiderman, S.R. Chatterjee, D., Chatterjee-Chakraborty, M., Rees, S., Cau- Tulkin, & A. Rosenfeld (Eds.), Culture and infancy: chi, J., de Medeiros, C.B., & Fleming, A.S. (2007). Variations in the human experience (pp. 151–188). New Maternal isolation alters the expression of neural York: Academic Press. proteins during development: ‘Stroking’ stimulation Bremner, J.D., Elzinga, B., Schmahl, C., & Vermetten, E. reverses these effects. Brain Research, 1158, 11–27. (2008). Structural and functional plasticity of the human Chaudron, L.H., Kitzman, H.J., Peifer, K.L., Morrow, S., brain in posttraumatic stress disorder. Progress in Brain Perez, L.M., & Newman, M.C. (2005). Self-recognition of Research, 167, 171–186. and provider response to maternal depressive symptoms Bridges, R. (2008). Neurobiology of the parental brain. in low-income Hispanic women. Journal of Women’s Amsterdam: Academic. Health, 14, 331–338. Bridges, R.S. (1990). Endocrine regulation of parental Chisholm, J.S. (1978). Swaddling, cradleboards and the behavior in rodents. In N.A. Krasnegor, & R.S. Bridges development of children. Early Human Development, 2, (Eds.), Mammalian parenting: Biochemical, neurobiologi- 255–275. cal, and behavioral determinants (pp. 93–117). New Christoff, K., Keramatian, K., Gordon, A.M., Smith, R., & York: Oxford University Press. Ma¨dler, B. (2009). Prefrontal organization of cognitive Brown, G.W., & Moran, P. (1994). Clinical and psychoso- control according to levels of abstraction. Brain Re- cial origins of chronic depressive episodes. I: A commu- search, 1286, 94–105. nity survey. British Journal of Psychiatry, 165, 447–456. Cicchetti, D., Rogosch, F.A., & Toth, S.L. (1998). Maternal Brown, J., Cohen, P., Johnson, J.G., & Smailes, E.M. depressive disorder and contextual risk: Contributions (1999). Childhood abuse and neglect: Specificity of to the development of attachment insecurity and behav- effects on adolescent and young adult depression and ior problems in toddlerhood. Development and Psycho- suicidality. Journal of the American Academy of Child pathology, 10, 283–300. and Adolescent Psychiatry, 38, 1490–1496. Clark, L., Chamberlain, S.R., & Sahakian, B.J. (2009). Brown, J.W., & Braver, T.S. (2005). Learned predictions of Neurocognitive mechanisms in depression: Implications error likelihood in the anterior cingulate cortex. Science, for treatment. Annual Review of Neuroscience, 32, 57– 307, 1118–1121. 74. Brown, J.W., & Braver, T.S. (2008). A computational model Cohn, J.F., Campbell, S.B., Matias, R., & Hopkins, J. of risk, conflict, and individual difference effects in the (1990). Face-to-face interactions of postpartum anterior cingulate cortex. Brain Research, 1202, 99–108. depressed and nondepressed mother–infant pairs at Buist, A. (2006). Perinatal depression – assessment and 2 months. Developmental Psychology, 35, 119–123. management. Australian Family Physician, 35, 670–673. Cole, M.W., & Schneider, W. (2007). The cognitive control Burton, C., Lovic, V., & Fleming, A.S. (2006). Early network: Integrated cortical regions with dissociable adversity alters attention and locomotion in adult functions. NeuroImage, 37, 343–360.

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. 390 Jennifer Barrett and Alison S. Fleming

Cooper, J.C., & Knutson, B. (2008). Valence and salience depression: An fMRI investigation. Journal of Affective contribute to nucleus accumbens activation. NeuroIm- Disorders, 114, 131–142. age, 39, 538–547. Dillon, D.G., Holmes, A.J., Birk, J.L., Brooks, N., Lyons- Cooper, P.J., & Murray, L. (1995). Course and recurrence Ruth, K., & Pizzagalli, D.A. (2009). Childhood adversity of postnatal depression. Evidence for the specificity of is associated with left basal ganglia dysfunction during the diagnostic concept. British Journal of Psychiatry, reward anticipation in adulthood. Biological Psychiatry, 166, 191–195. 66, 206–213. Coplan, J.D., Andrews, M.W., Rosenblum, L.A., Owens, Drevets, W.C., Price, J.L., & Furey, M.L. (2008). Brain M.J., Friedman, S., Gorman, J.M., et al. (1996). Persis- structural and functional abnormalities in mood disor- tent elevations of cerebrospinal fluid concentrations of ders: Implications for neurocircuitry models of depres- corticotropin-releasing factor in adult nonhuman pri- sion. Brain Structure and Function, 213, 93–118. mates exposed to early-life stressors: Implications for the Drevets, W.C., Savitz, J., & Trimble, M. (2008b). The pathophysiology of mood and anxiety disorders. Pro- subgenual anterior cingulate cortex in mood disorders. ceedings of the National Academy of Sciences of the CNS Spectrums, 13, 663–681. United States of America, 93, 1619–1623. Duncan, J., & Owen, A.M. (2000). Common regions of the Corter, C., & Fleming, A.S. (1990). Maternal responsive- human frontal lobe recruited by diverse cognitive ness in humans: Emotional, cognitive and biological demands. Trends in Neurosciences, 23, 475–483. factors. Advances in the Study of Behavior, 19, 83–136. Elliott, R., Sahakian, B.J., McKay, A.P., Herrod, J.J., Corter, C., & Fleming, A.S. (2002). Psychobiology of Robbins, T.W., & Paykel, E.S. (1996). Neuropsycholog- maternal behavior in human beings. In M.H. Bornstein ical impairments in unipolar depression: The influence (Ed.), Handbook of parenting: Biology and ecology of of perceived failure on subsequent performance. Psycho- parenting (pp. 141–182). Mahwah, NJ: Lawrence Erl- logical Medicine, 26, 975–989. baum Associates. Enns, M.W., Cox, B.J., & Clara, I. (2002). Parental bonding Cote, L.R., & Bornstein, M.H. (2009). Child and mother and adult psychopathology: Results from the US play in three US cultural groups: Comparisons and National Comorbidity Survey. Psychological Medicine, associations. Journal of Family Psychology, 23, 355– 32, 997–1008. 363. Fahlke, C., Lorenz, J.G., Long, J., Champoux, M., Suomi, Cox, J.L., Murray, D., & Chapman, G. (1993). A controlled S.J., & Higley, J.D. (2000). Rearing experiences and study of the onset, duration and prevalence of postnatal stress-induced plasma cortisol as early risk factors for depression. British Journal of Psychiatry, 163, 27–31. excessive alcohol consumption in nonhuman primates. Critchley, H.D. (2009). Psychophysiology of neural, cogni- Alcoholism, Clinical and Experimental Research, 24, 644– tive and affective integration: fMRI and autonomic 650. indicants. International Journal of Psychophysiology, Feldman, R., Granat, A., Pariente, C., Kanety, H., Kuint, J., 73, 88–94. & Gilboa-Schechtman, E. (2009). Maternal depression Dannlowski, U., Ohrmann, P., Bauer, J., Deckert, J., and anxiety across the postpartum year and infant social Hohoff, C., Kugel, H., et al. (2008). 5-HTTLPR biases engagement, fear regulation, and stress reactivity. Jour- amygdala activity in response to masked facial expres- nal of the American Academy of Child and Adolescent sions in major depression. Neuropsychopharmacology, Psychiatry, 48, 919–927. 33, 418–424. Feldman, R., Weller, A., Zagoory-Sharon, O., & Levine, A. Dannlowski, U., Ohrmann, P., Bauer, J., Kugel, H., Arolt, (2007). Evidence for a neuroendocrinological foundation V., Heindel, W., et al. (2007). Amygdala reactivity of human affiliation: Plasma oxytocin levels across to masked negative faces is associated with auto- pregnancy and the postpartum period predict mother– matic judgmental bias in major depression: A 3 T fMRI infant bonding. Psychological Science, 18, 965–970. study. Journal of Psychiatry and Neuroscience, 32, 423– Field, T. (1995). Psychologically depressed parents. In M. 429. Bernstein (Ed.), Handbook of parenting. Mahwah, NJ: Davey, C.G., Allen, N.B., Harrison, B.J., Dwyer, D.B., & Lawrence Erlbaum Associates. Yu¨cel, M. (2009). Being liked activates primary reward Field, T. (2009). Postpartum depression effects on early and midline self-related brain regions. Human Brain interactions, parenting, and safety practices: A review. Mapping, 31, 660–668. Infant Behavior and Development, 33, 1–6. De Bellis, M.D. (2005). The psychobiology of neglect. Child Field, T., Diego, M., Hernandez-Reif, M., Schanberg, S., & Maltreatment, 10, 150–172. Kuhn, C. (2002). Right frontal EEG and pregnancy/ Debiec, J. (2005). Peptides of love and fear: Vasopressin neonatal outcomes. Psychiatry, 65, 35–47. and oxytocin modulate the integration of information in Field, T., Healy, B., Goldstein, S., & Guthertz, M. (1990). the amygdala. BioEssays, 27, 869–873. Behavior-state matching and synchrony in mother– de Pau´ l, J. & Domenech, L. (2000). Childhood history of infant interactions of versus depressed dyads. Develop- abuse and child abuse potential in adolescent mothers: a mental Psychology, 26, 7–14. longitudinal study. Child Abuse & Neglect, 24(5), 701– Field, T., Hernandez-Reif, M., Diego, M., Feijo, L., Vera, Y., 713. Gil, K., et al. (2007). Still-face and separation effects on Derntl, B., Habel, U., Windischberger, C., Robinson, S., depressed mother–infant interactions. Infant Mental Kryspin-Exner, I., Gur, R.C., et al. (2009). General and Health Journal, 28, 314–323. specific responsiveness of the amygdala during explicit Field, T., Sandberg, D., Garcia, R., Bega-Lahr, N., Gold- emotion recognition in females and males. BMC Neuro- stein, S., & Guy, L. (1985). Pregnancy problems, post- science, 10, 91. partum depression, and early mother–infant Di Chiara, G. (2002). Nucleus accumbens shell and core interactions. Developmental Psychobiology, 21, 1152– dopamine: Differential role in behavior and addiction. 1156. Behavioural Brain Research, 137, 75–114. Figueira, P., Fernandes Malloy-Diniz, L., Aure´lio Romano- Dichter, G.S., Felder, J.N., & Smoski, M.J. (2009). Affective Silva, M., Silva Neves, F., & Correˆa, H. (2009). Post- context interferes with cognitive control in unipolar partum depression and comorbid disorders: Frequency

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. Neurobiological perspectives on mothering 391

and relevance to clinical management. Archives of Friedman, S.H., & Resnick, P.J. (2009). Postpartum Women’s Mental Health, 12, 451. depression: An update. Women’s Health, 5, 287–295. Fineberg, N.A., Potenza, M.N., Chamberlain, S.R., Berlin, Fu, C.H.Y., Williams, S.C.R., Cleare, A.J., Scott, J., H.A., Menzies, L., Bechara, A., Sahakian, B.J., et al. Mitterschiffthaler, M.T., Walsh, N.D., et al. (2008). (2010). Probing compulsive and impulsive behaviors, Neural responses to sad facial expressions in major from animal models to endophenotypes: A narrative depression following cognitive behavioral therapy. Bio- review. Neuropsychopharmacology, 35, 591–604. logical Psychiatry, 64, 505–512. Fishbein, D., Warner, T., Krebs, C., Trevarthen, N., Flan- Fujiwara, J., Tobler, P.N., Taira, M., Iijima, T., & Tsutsui, nery, B., & Hammond, J. (2009). Differential relation- K. (2009). Segregated and integrated coding of reward ships between personal and community stressors and and punishment in the cingulate cortex. Journal of children’s neurocognitive functioning. Child Maltreat- Neurophysiology, 101, 3284–3293. ment, 14, 299–315. Furstenberg, F.F., Brooks-Gunn, J., & Morgan, S.P. Fisher, H., Aron, A., & Brown, L.L. (2005). Romantic love: (1987). Adolescent mothers and their children in later An fMRI study of a neural mechanism for mate choice. life. Family Planning Perspectives, 19, 142–151. Journal of Comparative Neurology, 493, 58–62. Gale, S., & Harlow, B.L. (2003). Postpartum mood disor- Fleming, A.S., Corter, C., Franks, P., Surbey, M., Schnei- ders: A review of clinical and epidemiological factors. der, B., & Steiner, M. (1993). Postpartum factors related Journal of Psychosomatic Obstetrics and Gynaecology, to mother’s attraction to newborn infant odors. Develop- 24, 257–266. mental Psychobiology, 26, 115–132. Gamer, M., & Bu¨chel, C. (2009). Amygdala activation Fleming, A.S., & Korsmit, M. (1996). Plasticity in the predicts gaze toward fearful eyes. Journal of Neurosci- maternal circuit: Effects of maternal experience on Fos- ence, 29, 9123–9126. lir in hypothalamic, limbic and cortical structures in the Garner, B., Wood, S.J., Pantelis, C., & van den Buuse, M. postpartum rat. Behavioral Neuroscience, 110, 567–582. (2007). Early maternal deprivation reduces prepulse Fleming, A.S., Kraemer, G.W., Gonzalez, A., Lovic, V., Rees, inhibition and impairs spatial learning ability in adult- S., & Melo, A. (2002). Mothering begets mothering: The hood: No further effect of post-pubertal chronic cortico- transmission of behavior and its neurobiology across sterone treatment. Behavioural Brain Research, 176, generations. Pharmacology, Biochemistry, and Behavior, 323–332. 73, 61–75. Gavin, N.I., Gaynes, B.N., Lohr, K.N., Meltzer-Brody, S., Fleming, A.S., & Rosenblatt, J.S. (1974a). Olfactory regu- Gartlehner, G., & Swinson, T. (2005). Perinatal depres- lation of maternal behavior in rats. I. Effects of olfactory sion: A systematic review of prevalence and incidence. bulb removal in experienced and inexperienced lactating Obstetrics and Gynecology, 106(5 Pt 1), 1071–1083. and cycling females. Journal of Comparative and Phys- Giacobbe, P., Mayberg, H.S., & Lozano, A.M. (2009). iological Psychology, 86, 221–232. Treatment resistant depression as a failure of brain Fleming, A.S., & Rosenblatt, J.S. (1974b). Olfactory regu- homeostatic mechanisms: Implications for deep brain lation of maternal behavior in rats. II. Effects of periph- stimulation. Experimental Neurology, 219, 44–52. erally induced anosmia and lesions of the lateral Giardino, J., Gonzalez, A., Steiner, M., & Fleming, A.S. olfactory tract in pup-induced virgins. Journal of Com- (2008). Effects of motherhood on physiological and parative and Physiological Psychology, 86, 233–246. subjective responses to infant cries in teenage mothers: Fleming, A.S., Ruble, D.N., Flett, G.L., & Shaul, D. (1988). A comparison with non-mothers and adult mothers. Postpartum adjustment in first-time mothers: Relations Hormones and Behavior, 53, 149–158. between mood, maternal attitudes and mother–infant Glocker, M.L., Langleben, D.D., Ruparel, K., Loughead, interactions. Developmental Psychology, 24, 71–81. J.W., Valdez, J.N., Griffin, M.D., et al. (2009). Baby Fleming, A.S., Ruble, D., Krieger, H., & Wong, P.Y. (1997a). schema modulates the brain reward system in nullipa- Hormonal and experiential correlates of maternal rous women. Proceedings of the National Academy of responsiveness during pregnancy and the puerperium Sciences of the United States of America, 106, 9115–9119. in human mothers. Hormones and Behavior, 31, 145– Gobbini, M.I., & Haxby, J.V. (2007). Neural systems for 158. recognition of familiar faces. Neuropsychologia, 45,32–41. Fleming, A.S., Steiner, M., & Anderson, V. (1987). Hor- Goldstone, A.P., de Hernandez, C.G.P., Beaver, J.D., monal and attitudinal correlates of maternal behavior Muhammed, K., Croese, C., Bell, G., et al. (2009). during the early postparpregnancy. Journal of Reproduc- Fasting biases brain reward systems towards high- tive and Infant Psychology, 5, 193–205. calorie foods. European Journal of Neuroscience, 30, Fleming, A.S., Steiner, M., & Corter, C. (1997b). Cortisol, 1625–1635. hedonics, and maternal responsiveness in human moth- Gonzalez, A., & Fleming, A.S. (2002). Artificial rearing ers. Hormones and Behavior, 32, 85–98. causes changes in maternal behavior and c-fos expres- Fossati, P., Ergis, A.M., & Allilaire, J.F. (2001). Problem- sion in juvenile female rats. Behavioral Neuroscience, solving abilities in unipolar depressed patients: Com- 116, 999–1013. parison of performance on the modified version of the Gonzalez, A., Jenkins, J.M., Steiner, M., & Fleming, A.S. Wisconsin and the California sorting tests. Psychiatry (2009). The relation between early life adversity, cortisol Research, 104, 145–156. awakening response and diurnal salivary cortisol levels France, R.D., Urban, B., Krishnan, K.R., Bissette, G., in postpartum women. Psychoneuroendocrinology, 34, Banki, C.M., Nemeroff, C., et al. (1988). CSF corticotro- 76–86. pin-releasing factor-like immunoactivity in chronic pain Gonzalez, A., Lovic, V., Ward, G.R., Wainwright, P.E., & patients with and without major depression. Biological Fleming, A.S. (2001). Intergenerational effects of com- Psychiatry, 23, 86–88. plete maternal deprivation and replacement stimulation Francis, D., Diorio, J., Liu, D., & Meaney, M.J. (1999). on maternal behavior and emotionality in female rats. Nongenomic transmission across generations of mater- Developmental Psychobiology, 38, 11–32. nal behavior and stress responses in the rat. Science, Goto, Y., & Grace, A.A. (2005). Dopamine-dependent 286, 1155–1158. interactions between limbic and prefrontal cortical plas-

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. 392 Jennifer Barrett and Alison S. Fleming

ticity in the nucleus accumbens: Disruption by cocaine hood: Tests of developmental pathways involving ado- sensitization. Neuron, 47, 255–266. lescent risk taking and depression. Development and Goto, Y., & Grace, A.A. (2008). Limbic and cortical infor- Psychopathology, 22, 655–665. mation processing in the nucleus accumbens. Trends in Higley, J.D., Suomi, S.J., & Linnoila, M. (1991). CSF Neurosciences, 31, 552–558. monoamine metabolite concentrations vary according to Goyal, D., Gay, C., & Lee, K. (2009). Fragmented maternal age, rearing, and sex, and are influenced by the stressor sleep is more strongly correlated with depressive symp- of social separation in rhesus monkeys. Psychopharma- toms than infant temperament at three months postpar- cology, 103, 551–556. tum. Archives of Women’s Mental Health, 12, 229–237. Higley, J.D., Suomi, S.J., & Linnoila, M. (1992). A longi- Grabenhorst, F., Rolls, E.T., Parris, B.A., & d’Souza, A.A. tudinal assessment of CSF monoamine metabolite and (2009). How the brain represents the reward value of fat plasma cortisol concentrations in young rhesus mon- in the mouth. Cerebral Cortex, 20, 1082–1091. keys. Biological Psychiatry, 32, 127–145. Grant, M.M., Thase, M.E., & Sweeney, J.A. (2001). Cogni- Hollerman, J.R., Tremblay, L., & Schultz, W. (2000). tive disturbance in outpatient depressed younger adults: Involvement of basal ganglia and orbitofrontal cortex in Evidence of modest impairment. Biological Psychiatry, goal-directed behavior. Progress in Brain Research, 126, 50, 35–43. 193–215. Gray, J., Venn, H., Montagne, B., Murray, L., Burt, M., Horn, N.R., Dolan, M., Elliott, R., Deakin, J.F.W., & Frigerio, E., et al. (2006). Bipolar patients show mood- Woodruff, P.W.R. (2003). Response inhibition and congruent biases in sensitivity to facial expressions of impulsivity: An fMRI study. Neuropsychologia, 41, emotion when exhibiting depressed symptoms, but not 1959–1966. when exhibiting manic symptoms. Cognitive Neuropsy- Hrdy, S. (2009). Mothers and others: The evolutionary chiatry, 11, 505–520. origins of mutual understanding. Cambridge MA: Bel- Groe¨r, M.W., & Morgan, K. (2007). Immune, health and knap Press of Harvard University Press. endocrine characteristics of depressed postpartum Hrdy, S.B. (2005). Evolutionary context of human devel- mothers. Psychoneuroendocrinology, 32, 133–139. opment: The cooperative breeding model. In C.S. Carter, Gump, B.B., Reihman, J., Stewart, P., Lonky, E., Darvill, L. Anhert, K. E. Grossmann, S.B. Hrdy, M.E. Lamb, S.W. T., Granger, D.A., et al. (2009). Trajectories of maternal Porges, & N. Sachser (Eds.), Attachment and bonding: A depressive symptoms over her child’s life span: Relation new synthesis (pp. 9–32). Cambridge, MA: MIT Press. To adrenocortical, cardiovascular, and emotional func- Huber, D., Veinante, P., & Stoop, R. (2005). Vasopressin tioning in children. Development and Psychopathology, and oxytocin excite distinct neuronal populations in the 21, 207–225. central amygdala. Science), 308, 245–248. Gu¨ngo¨r, D., & Bornstein, M.H. (in press). Culture-general Insel, T. (1990). Oxytocin and maternal behavior. In N.A. and -specific associations of attachment avoidance and Krasnegor, & R.S. Bridges (Eds.), Mammalian parenting: anxiety with perceived parental warmth and psycholog- Biochemical, neurobiological, and behavioral determi- ical control among Turk and Belgian adolescents. Jour- nants (pp. 260–280). New York: Oxford University Press. nal of Adolescence. Inoue, Y., Tonooka, Y., Yamada, K., & Kanba, S. (2004). Gunnar, M.R., Brodersen, L., Nachmias, M., Buss, K., & Deficiency of theory of mind in patients with remitted Rigatuso, J. (1996). Stress reactivity and attachment mood disorder. Journal of Affective Disorders, 82, 403– security. Developmental Psychobiology, 29, 191–204. 409. Gunnar, M.R., & Fisher, P.A. (2006). Bringing basic Inoue, Y., Yamada, K., & Kanba, S. (2006). Deficit in theory research on early experience and stress neurobiology to of mind is a risk for relapse of major depression. Journal bear on preventive interventions for neglected and mal- of Affective Disorders, 95, 125–127. treated children. Development and Psychopathology, 18, Jaffee, S.R., Caspi, A., Moffitt, T.E., Polo-Toma´s, M., & 651–677. Taylor, A. (2007). Individual, family, and neighborhood Halligan, S.L., Murray, L., Martins, C., & Cooper, P.J. factors distinguish resilient from non-resilient mal- (2007). Maternal depression and psychiatric outcomes in treated children: A cumulative stressors model. Child adolescent offspring: A 13-year longitudinal study. Jour- Abuse and Neglect, 31, 231–253. nal of Affective Disorders, 97, 145–154. Johansen-Berg, H., Gutman, D.A., Behrens, T E.J., Mat- Harkness, K.L., & Monroe, S.M. (2002). Childhood adver- thews, P.M., Rushworth, M.F.S., Katz, E., et al. (2008). sity and the endogenous versus nonendogenous distinc- Anatomical connectivity of the subgenual cingulate region tion in women with major depression. American Journal targeted with deep brain stimulation for treatment-resis- of Psychiatry, 159, 387–393. tant depression. Cerebral Cortex, 18, 1374–1383. Heim, C., & Nemeroff, C.B. (1999). The impact of early Jolley, S.N., Elmore, S., Barnard, K.E., & Carr, D.B. adverse experiences on brain systems involved in the (2007). Dysregulation of the hypothalamic-pituitary- pathophysiology of anxiety and affective disorders. Bio- adrenal axis in postpartum depression. Biological logical Psychiatry, 46, 1509–1522. Research for Nursing, 8, 210–222. Heim, C., Plotsky, P.M., & Nemeroff, C.B. (2004). Impor- Kammerer, M., Taylor, A., & Glover, V. (2006). The HPA tance of studying the contributions of early adverse axis and perinatal depression: A hypothesis. Archives of experience to neurobiological findings in depression. Women’s Mental Health, 9, 187–196. Neuropsychopharmacology, 29, 641–648. Kaneda, Y. (2009). Verbal working memory impairment in Hercher, C., Turecki, G., & Mechawar, N. (2009). Through patients with current episode of unipolar major depres- the looking glass: Examining neuroanatomical evidence sive disorder and in remission. Clinical Neuropharmacol- for cellular alterations in major depression. Journal of ogy, 32, 346–347. Psychiatric Research, 43, 947–961. Kannan, S., Carruth, B.R., & Skinner, J. (2004). Neonatal Herrenkohl, T.I., Kosterman, R., Mason, W.A., Hawkins, feeding practices of Anglo American mothers and Asian J.D., McCarty, C.A., & McCauley, E. (2010). Effects of Indian mothers living in the United States and India. childhood conduct problems and family adversity on Journal of Nutrition Education and Behavior, 36, 315– health, health behaviors, and service use in early adult- 319.

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. Neurobiological perspectives on mothering 393

Karevold, E., Røysamb, E., Ystrom, E., & Mathiesen, K.S. ment security: Investigating the mediating role of mater- (2009). Predictors and pathways from infancy to symp- nal sensitivity. Infant Behavior and Development, 31, toms of anxiety and depression in early adolescence. 688–695. Developmental Psychology, 45, 1051–1060. Leh, S.E., Petrides, M., & Strafella, A.P. (2010). The neural Kaufman, J., Plotsky, P.M., Nemeroff, C.B., & Charney, circuitry of executive functions in healthy subjects and D.S. (2000). Effects of early adverse experiences on brain Parkinson’s disease. Neuropsychopharmacology, 35,70–85. structure and function: Clinical implications. Biological Leibenluft, E., Gobbini, M.I., Harrison, T., & Haxby, J.V. Psychiatry, 48, 778–790. (2004). Mothers’ neural activation in response to pic- Kaufman, J., Yang, B., Douglas-Palumberi, H., Houshyar, tures of their children and other children. Biological S., Lipschitz, D., Krystal, J.H., et al. (2004). Social Psychiatry, 56, 225–232. supports and serotonin transporter gene moderate Leiderman, P., & Leiderman, M. (1977). Economic change depression in maltreated children. Proceedings of the and infant care in an East African agricultural commu- National Academy of Sciences of the United States of nity. In P.H. Leiderman, S.R. Tulkin, & A. Rosenfeld America, 101, 17316–17321. (Eds.), Culture and infancy: Variations in human experi- Kauppi, A., Kumpulainen, K., Vanamo, T., Merikanto, J., & ence (pp. 405–438). New York: Academic Press. Karkola, K. (2008). Maternal depression and filicide-case Leifer, M. (1980). Psychological effects of motherhood: A study of ten mothers. Archives of Women’s Mental study of first pregnancy. New York: Praeger. Health, 11, 201–206. Levitan, R.D., Rector, N.A., Sheldon, T., & Goering, P. Keverne, E.B., & Curley, J.P. (2004). Vasopressin, oxytocin (2003). Childhood adversities associated with major and social behaviour. Current Opinion in Neurobiology, depression and/or anxiety disorders in a community 14, 777–783. sample of Ontario: Issues of co-morbidity and specificity. Killgore, W.D.S., & Yurgelun-Todd, D.A. (2004). Activation Depression and Anxiety, 17, 34–42. of the amygdala and anterior cingulate during noncon- Li, X., Lu, Z., D’Argembeau, A., Ng, M., & Bechara, A. scious processing of sad versus happy faces. NeuroIm- (2009). The Iowa Gambling Task in fMRI images. Human age, 21, 1215–1223. Brain Mapping, 31, 410–423. Kim, J., Kim, S., Kim, J., Jeong, B., Park, C., Son, A.R., Lizardi, H., Klein, D.N., Ouimette, P.C., Riso, L.P., Ander- et al. (2009a). Compassionate attitude towards others’ son, R.L., & Donaldson, S.K. (1995). Reports of the suffering activates the mesolimbic neural system. Neu- childhood home environment in early-onset dysthymia ropsychologia, 47, 2073–2081. and episodic major depression. Journal of Abnormal Kim, P., Leckman, J.F., Mayes, L.C., Newman, M., Feld- Psychology, 104, 132–139. man, R., & Swain, J.E. (2009b). Perceived quality of Lorberbaum, J.P., Newman, J.D., Horwitz, A.R., Dubno, maternal care in childhood and structure and function of J.R., Lydiard, R.B., Hamner, M.B., et al. (2002). A mothers’ brain. Developmental Science, 13, 662–673. potential role for thalamocingulate circuitry in human Kirsch, P., Esslinger, C., Chen, Q., Mier, D., Lis, S., maternal behavior. Biological Psychiatry, 51, 431–445. Siddhanti, S., et al. (2005). Oxytocin modulates neural Lovic, V., & Fleming, A.S. (2004). Artificially-reared female circuitry for social cognition and fear in humans. Journal rats show reduced prepulse inhibition and deficits in the of Neuroscience, 25, 11489–11493. attentional set shifting task – reversal of effects with Klein, D.N., Arnow, B.A., Barkin, J.L., Dowling, F., Kocsis, maternal-like licking stimulation. Behavioural Brain J.H., Leon, A.C., et al. (2009). Early adversity in chronic Research, 148, 209–219. depression: Clinical correlates and response to pharma- Lovic, V., Gonzalez, A., & Fleming, A.S. (2001). Maternally cotherapy. Depression and Anxiety, 26, 701–710. separated rats show deficits in maternal care in adult- Knutson, J.F. (1995). Psychological characteristics of mal- hood. Developmental Psychobiology, 39, 19–33. treated children: Putative risk factors and consequences. Maestripieri, D. (2005). Early experience affects the inter- Annual Review of Psychology, 46, 401–431. generational transmission of infant abuse in rhesus Koenigs, M., & Grafman, J. (2009). The functional neuro- monkeys. Proceedings of the National Academy of Sci- anatomy of depression: Distinct roles for ventromedial ences of the United States of America, 102, 9726–9729. and dorsolateral prefrontal cortex. Behavioural Brain Marcus, S.M. (2009). Depression during pregnancy: Rates, Research, 201, 239–243. risks and consequences – Motherisk Update 2008. Koo, V., Lynch, J., & Cooper, S. (2003). Risk of postnatal Canadian Journal of Clinical Pharmacology, 16, e15–22. depression after emergency delivery. Journal of Obstet- Martin, G., Bergen, H.A., Roeger, L., & Allison, S. (2004). rics and Gynaecology Research, 29, 246–250. Depression in young adolescents: Investigations using 2 Kreppner, J.M., O’Connor, T.G., & Rutter, M. (2001). Can and 3 factor versions of the Parental Bonding Instrument. inattention/overactivity be an institutional deprivation Journal of Nervous and Mental Disease, 192, 650–657. syndrome? Journal of Abnormal Child Psychology, 29, Matsuo, K., Nicoletti, M., Nemoto, K., Hatch, J.P., Peluso, 513–528. M.A.M., Nery, F.G., et al. (2009). A voxel-based mor- Kringelbach, M.L. (2005). The human orbitofrontal cortex: phometry study of frontal gray matter correlates of Linking reward to hedonic experience. Nature Reviews. impulsivity. Human Brain Mapping, 30, 1188–1195. Neuroscience, 6, 691–702. Mattson, B.J., & Morrell, J.I. (2005). Preference for Krpan, K.M., Coombs, R., Zinga, D., Steiner, M., & Flem- cocaine- versus pup-associated cues differentially acti- ing, A.S. (2005). Experiential and hormonal correlates of vates neurons expressing either Fos or cocaine- and maternal behavior in teen and adult mothers. Hormones amphetamine-regulated transcript in lactating, maternal and Behavior, 47, 112–122. rodents. Neuroscience, 135, 315–328. Krumbholz, K., Nobis, E.A., Weatheritt, R.J., & Fink, G.R. Mayberg, H.S., Lozano, A.M., Voon, V., McNeely, H.E., (2009). Executive control of spatial attention shifts in the Seminowicz, D., Hamani, C., et al. (2005). Deep brain auditory compared to the visual modality. Human Brain stimulation for treatment-resistant depression. Neuron, Mapping, 30, 1457–1469. 45, 651–660. Laranjo, J., Bernier, A., & Meins, E. (2008). Associations McCauley, J., Kern, D.E., Kolodner, K., Dill, L., Schroeder, between maternal mind-mindedness and infant attach- A.F., DeChant, H.K., et al. (1997). Clinical characteris-

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. 394 Jennifer Barrett and Alison S. Fleming

tics of women with a history of childhood abuse: Journal of the American Medical Women’s Association, Unhealed wounds. Journal of the American Medical 59, 181–191. Association, 277, 1362–1368. Moskvina, V., Farmer, A., Swainson, V., O’Leary, J., McEwen, B.S. (2003). Early life influences on life-long patterns Gunasinghe, C., Owen, M., et al. (2007). Interrelation- of behavior and health. Mental Retardation and Develop- ship of childhood trauma, neuroticism, and depressive mental Disabilities Research Reviews, 9, 149–154 4. phenotype. Depression and Anxiety, 24, 163–168. Meins, E., Fernyhough, C., Fradley, E., & Tuckey, M. Moss, H.A., & Jones, S.J. (1977). Relations between (2001). Rethinking maternal sensitivity: Mothers’ com- maternal attitudes and maternal behavior as a function ments on infants’ mental processes predict security of of social class. In P.H. Leiderman, S.R. Tulkin, & A. attachment at 12 months. Journal of Child Psychology Rosenfeld (Eds.), Culture and infancy. New York: Aca- and Psychiatry, 42, 637–648. demic Press. Meins, E., Fernyhough, C., Russell, J., & Clark-Carter, D. Murray, L., Woolgar, M., Murray, J., & Cooper, P. (2003). (1998). Security of attachment as a predictor of symbolic Self-exclusion from health care in women at high risk for and mentalising abilities: A longitudinal study. Social postpartum depression. Journal of Public Health Medi- Development, 7, 1–24. cine, 25, 131–137. Meins, E., Fernyhough, C., Wainwright, R., Clark-Carter, Nachmias, M., Gunnar, M., Mangelsdorf, S., Parritz, R.H., D., Das Gupta, M., Fradley, E., et al. (2003). Pathways to & Buss, K. (1996). Behavioral inhibition and stress understanding mind: Construct validity and predictive reactivity: The moderating role of attachment security. validity of maternal mind-mindedness. Child Develop- Child Development, 67, 508–522. ment, 74, 1194–1211. Nagahama, Y., Okada, T., Katsumi, Y., Hayashi, T., Meins, E., Fernyhough, C., Wainwright, R., Das Gupta, M., Yamauchi, H., Oyanagi, C., et al. (2001). Dissociable Fradley, E., & Tuckey, M. (2002). Maternal mind-mind- mechanisms of attentional control within the human edness and attachment security as predictors of theory prefrontal cortex. Cerebral Cortex, 11, 85–92. of mind understanding. Child Development, 73, 1715– Nakano, Y., Baba, H., Maeshima, H., Kitajima, A., Sakai, 1726. Y., Baba, K., Suzuki, T., et al. (2008). Executive dys- Melo, A.I., Lovic, V., Gonzalez, A., Madden, M., Sinopoli, K., function in medicated, remitted state of major depres- & Fleming, A.S. (2006). Maternal and littermate depri- sion. Journal of Affective Disorders, 111, 46–51. vation disrupts maternal behavior and social-learning of Nee, D.E., & Jonides, J. (2009). Common and distinct food preference in adulthood: Tactile stimulation, nest neural correlates of perceptual and memorial selection. odor, and social rearing prevent these effects. Develop- NeuroImage, 45, 963–975. mental Psychobiology, 48, 209–219. Nelson, E.E., Herman, K.N., Barrett, C.E., Noble, P.L., Milad, M.R., & Rauch, S.L. (2007). The role of the orbito- Wojteczko, K., Chisholm, K., et al. (2009). Adverse frontal cortex in anxiety disorders. Annals of the New rearing experiences enhance responding to both aversive York Academy of Sciences, 1121, 546–561. and rewarding stimuli in juvenile rhesus monkeys. Miller, L., Kramer, R., Warner, V., Wickramaratne, P., & Biological Psychiatry, 66, 702–704. Weissman, M. (1997). Intergenerational transmission of Nemeroff, C.B., Bissette, G., Akil, H., & Fink, M. (1991). parental bonding among women. Journal of the American Neuropeptide concentrations in the cerebrospinal fluid of Academy of Child and Adolescent Psychiatry, 36, 1134– depressed patients treated with electroconvulsive ther- 1139. apy. Corticotrophin-releasing factor, beta-endorphin and Miller, L., Warner, V., Wickramaratne, P., & Weissman, M. somatostatin. British Journal of Psychiatry, 158, 59–63. (1999). Self-esteem and depression: Ten year follow-up Nemeroff, C.B., Widerlo¨v, E., Bissette, G., Walle´us, H., of mothers and offspring. Journal of Affective Disorders, Karlsson, I., Eklund, K., et al. (1984). Elevated concen- 52, 41–49. trations of CSF corticotropin-releasing factor-like Mitchell, J.P., Neil Macrae, C., & Banaji, M.R. (2005). immunoreactivity in depressed patients. Science, 226, Forming impressions of people versus inanimate objects: 1342–1344. Social-cognitive processing in the medial prefrontal Nierop, A., Bratsikas, A., Zimmermann, R., & Ehlert, U. cortex. NeuroImage, 26, 251–257. (2006). Are stress-induced cortisol changes during preg- Modirrousta, M., & Fellows, L.K. (2008). Dorsal medial nancy associated with postpartum depressive symp- prefrontal cortex plays a necessary role in rapid error toms? Psychosomatic Medicine, 68, 931–937. prediction in humans. Journal of Neuroscience, 28, Nitschke, J.B., Nelson, E.E., Rusch, B.D., Fox, A.S., 14000–14005. Oakes, T.R., & Davidson, R.J. (2004). Orbitofrontal Moehler, E., Biringen, Z., & Poustka, L. (2007). Emotional cortex tracks positive mood in mothers viewing pictures availability in a sample of mothers with a history of of their newborn infants. NeuroImage, 21, 583–592. abuse. American Journal of Orthopsychiatry, 77, 624– Nitschke, J. B., Sarinopoulos, I., Oathes, D. J., Johnstone, 628. T., Whalen, P. J., Davidson, R. J., et al. (2009). Anticipa- Monchi, O., Petrides, M., Petre, V., Worsley, K., & Dagher, tory activation in the amygdala and anterior cingulate in A. (2001). Wisconsin Card Sorting revisited: Distinct generalized anxiety disorder and prediction of treatment neural circuits participating in different stages of the response. American Journal of Psychiatry, 166, 302–310. task identified by event-related functional magnetic Nolin, P., & Ethier, L. (2007). Using neuropsychological resonance imaging. Journal of Neuroscience, 21, 7733– profiles to classify neglected children with or without 7741. physical abuse. Child Abuse and Neglect, 31, 631–643. Moritz, S., Birkner, C., Kloss, M., Jahn, H., Hand, I., Noriuchi, M., Kikuchi, Y., & Senoo, A. (2008). The func- Haasen, C., & Krausz, M. (2002). Executive functioning tional neuroanatomy of maternal love: Mother’s response in obsessive-compulsive disorder, unipolar depression, to infant’s attachment behaviors. Biological Psychiatry, and schizophrenia. Archives of Clinical Neuropsychology, 63, 415–423. 17, 477–483. Numan, M. (2007). Motivational systems and the neural Moses-Kolko, E.L., & Roth, E.K. (2004). Antepartum and circuitry of maternal behavior in the rat. Developmental postpartum depression: Healthy mom, healthy baby. Psychobiology, 49, 12–21.

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. Neurobiological perspectives on mothering 395

Numan, M., Fleming, A.S., & Levy, F. (2006). Maternal Platek, S.M., & Kemp, S.M. (2009). Is family special to the behavior. In J.D. Neill (Ed.), Knobil and Neill’s physiology brain? An event-related fMRI study of familiar, familial, of reproduction (pp. 1921–1993). San Diego, CA: Elsevier. and self-face recognition. Neuropsychologia, 47, 849–858. Numan, M., & Insel, T. (2003). The neurobiology of parental Pryce, C.R., Dettling, A., Spengler, M., Spaete, C., & behavior. New York: Springer. Feldon, J. (2004). Evidence for altered monoamine O’Hara, M.W., Zekoski, E.M., Philipps, L.H., & Wright, E.J. activity and emotional and cognitive disturbance in (1990). Controlled prospective study of postpartum marmoset monkeys exposed to early life stress. Annals mood disorders: Comparison of childbearing and non- of the New York Academy of Sciences, 1032, 245–249. childbearing women. Journal of Abnormal Psychology, Pryce, C.R., Martin, RD., & Skuse, D. (1995). Motherhood 99, 3–15. in human and nonhuman primates: Biosocial determi- Olaza´bal, D.E., Abercrombie, E., Rosenblatt, J.S., & Mor- nants. New York: Karger. rell, J.I. (2004). The content of dopamine, serotonin, and Putallaz, M., Costanzo, P.R., Grimes, C.L., & Sherman, their metabolites in the neural circuit that mediates D.M. (1998). Intergenerational continuities and their maternal behavior in juvenile and adult rats. Brain influences on children’s social development. Social Research Bulletin, 63, 259–268. Development, 7, 389–427. Palombo, D.J., Nowoslawski, M., & Fleming, A.S. (2010). Putnam, F.W. (2003). Ten-year research update review: Motherless rats show deficits in maternal behavior Child sexual abuse. Journal of the American Academy of towards fostered pups. Developmental Psychobiology, Child and Adolescent Psychiatry, 42, 269–278. 52, 142–148. Raes, F., Hermans, D., & Williams, J.M.G. (2006). Negative Pak-Gorstein, S., Haq, A., & Graham, E.A. (2009). Cultural bias in the perception of others’ facial emotional expres- influences on infant feeding practices. Pediatrics in sions in major depression: The role of depressive rumi- Review, 30, e11–21. nation. Journal of Nervous and Mental Disease, 194, Panthangi, V., West, P., Savoy-Moore, R.T., Geeta, M., & 796–799. Reickert, E. (2009). Is seasonal variation another risk Ranote, S., Elliott, R., Abel, K.M., Mitchell, R., Deakin, factor for postpartum depression? Journal of the Amer- J.F.W., & Appleby, L. (2004). The neural basis of ican Board of Family Medicine, 22, 492–497. maternal responsiveness to infants: An fMRI study. Paris, R., Bolton, R.E., & Weinberg, M.K. (2009). Neuroreport, 15, 1825–1829. Postpartum depression, suicidality, and mother–infant Rao, U., Hammen, C., Ortiz, L.R., Chen, L., & Poland, R.E. interactions. Archives of Women’s Mental Health, 12, (2008). Effects of early and recent adverse experiences 309–321. on adrenal response to psychosocial stress in depressed de Pau´ l, J., & Domenech, L. (2000). Childhood history of adolescents. Biological Psychiatry, 64, 521–526. abuse and child abuse potential in adolescent mothers: Rauch, S.L., Shin, L.M., & Phelps, E.A. (2006). Neurocir- A longitudinal study. Child Abuse and Neglect, 24, 701– cuitry models of posttraumatic stress disorder and 713. extinction: Human neuroimaging research – past, pres- Paus, T. (2001). Primate anterior cingulate cortex: Where ent, and future. Biological Psychiatry, 60, 376–382. motor control, drive and cognition interface. Nature Ravizza, S.M., & Carter, C.S. (2008). Shifting set about Reviews. Neuroscience, 2, 417–424. task switching: Behavioral and neural evidence for Pears, K., & Fisher, P.A. (2005). Developmental, cognitive, distinct forms of cognitive flexibility. Neuropsychologia, and neuropsychological functioning in preschool-aged 46, 2924–2935. foster children: Associations with prior maltreatment Rebollo, A.G., & Montero, C.M. (2000). Variables perina- and placement history. Journal of Developmental and tales y desigualdades en salud en un a´rea sanitaria de Behavioral Pediatrics, 26, 112–122. Ca´ceres [Perinatal variables and health inequalities in a Pears, K.C., Kim, H.K., & Fisher, P.A. (2008). Psychosocial health-care district in Ca´ceres, Spain]. Gaceta Sanitaria, and cognitive functioning of children with specific pro- 14, 31–38. files of maltreatment. Child Abuse and Neglect, 32, 958– Rees, S.L., Akbari, E., Steiner, M., & Fleming, A.S. (2008). 971. Effects of early deprivation and maternal separation on Pederson, D.R., Moran, G., Sitko, C., Campbell, K., Ghes- pup-directed behavior and HPA axis measures in the quire, K., & Acton, H. (1990). Maternal sensitivity and juvenile female rat. Developmental Psychobiology, 50, the security of infant–mother attachment: A Q-sort 315–321. study. Child Development, 61, 1974–1983. Rees, S.L., Panesar, S., Steiner, M., & Fleming, A.S. (2004). Penza, K.M., Heim, C., & Nemeroff, C.B. (2003). Neurobi- The effects of adrenalectomy and corticosterone replace- ological effects of childhood abuse: Implications for the ment on maternal behavior in the postpartum rat. pathophysiology of depression and anxiety. Archives of Hormones and Behavior, 46, 411–419. Women’s Mental Health, 6, 15–22. Rees, S.L., Panesar, S., Steiner, M., & Fleming, A.S. (2006). Petrides, M. (2005). Lateral prefrontal cortex: Architectonic The effects of adrenalectomy and corticosterone replace- and functional organization. Philosophical Transactions ment on induction of maternal behavior in the virgin of the Royal Society of London. Series B, Biological female rat. Hormones and Behavior, 49, 337–345. Sciences, 360, 781–795. Repetti, R.L., Taylor, S.E., & Seeman, T.E. (2002). Risky Petrides, M., & Pandya, D.N. (1999). Dorsolateral prefron- families: Family social environments and the mental and tal cortex: Comparative cytoarchitectonic analysis in the physical health of offspring. Psychological Bulletin, 128, human and the macaque brain and corticocortical 330–366. connection patterns. European Journal of Neuroscience, Rey, J.M. (1995). Perceptions of poor maternal care are 11, 1011–1036. associated with adolescent depression. Journal of Affec- Pizzagalli, D.A., Holmes, A.J., Dillon, D.G., Goetz, E.L., tive Disorders, 34, 95–100. Birk, J.L., Bogdan, R., et al. (2009). Reduced caudate Righetti-Veltema, M., Conne-Perre´ard, E., Bousquet, A., & and nucleus accumbens response to rewards in unmed- Manzano, J. (2002). Postpartum depression and mother– icated individuals with major depressive disorder. Amer- infant relationship at 3 months old. Journal of Affective ican Journal of Psychiatry, 166, 702–710. Disorders, 70, 291–306.

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. 396 Jennifer Barrett and Alison S. Fleming

Rikhye, K., Tyrka, A.R., Kelly, M.M., Gagne, G.G., Mello, Schoenbaum, G., Roesch, M.R., Stalnaker, T.A., & Takah- A.F., Mello, M.F., et al. (2008). Interplay between child- ashi, Y.K. (2009). A new perspective on the role of the hood maltreatment, parental bonding, and gender orbitofrontal cortex in adaptive behaviour. Nature effects: Impact on quality of life. Child Abuse and Neglect, Reviews. Neuroscience, 10, 885–892. 32, 19–34. Schorscher-Petcu, A., Dupre´, A., & Tribollet, E. (2009). Risold, P.Y., Thompson, R.H., & Swanson, L.W. (1997). The Distribution of vasopressin and oxytocin binding sites in structural organization of connections between hypo- the brain and upper spinal cord of the common marmo- thalamus and cerebral cortex. Brain Research. Brain set. Neuroscience Letters, 461, 217–222. Research Reviews, 24, 197–254. Schulte-Ru¨ther, M., Markowitsch, H.J., Fink, G.R., & Rizzolatti, G., & Craighero, L. (2004). The mirror-neuron Piefke, M. (2007). Mirror neuron and theory of mind system. Annual Review of Neuroscience, 27, 169–192. mechanisms involved in face-to-face interactions: A Rizzolatti, G., & Fabbri-Destro, M. (2008). The mirror functional magnetic resonance imaging approach to system and its role in social cognition. Current Opinion empathy. Journal of Cognitive Neuroscience, 19, 1354– in Neurobiology, 18, 179–184. 1372. Roceri, M., Cirulli, F., Pessina, C., Peretto, P., Racagni, G., Schulz, K.P., Clerkin, S.M., Halperin, J.M., Newcorn, J.H., & Riva, M.A. (2004). Postnatal repeated maternal depri- Tang, C.Y., & Fan, J. (2009). Dissociable neural effects of vation produces age-dependent changes of brain-derived stimulus valence and preceding context during the neurotrophic factor expression in selected rat brain inhibition of responses to emotional faces. Human Brain regions. Biological Psychiatry, 55, 708–714. Mapping, 30, 2821–2833. Robson, K.S. (1967). The role of eye-to-eye contact in Seeman, T.E., Singer, B.H., Ryff, C.D., Dienberg Love, G., maternal–infant attachment. Journal of Child Psychology & Levy-Storms, L. (2002). Social relationships, gender, and Psychiatry, 87, 13–25. and allostatic load across two age cohorts. Psychoso- Robson, K.S., & Kumar, R. (1980). Delayed onset of matic Medicine, 64, 395–406. maternal affection after childbirth. British Journal of Seifritz, E., Esposito, F., Neuhoff, J.G., Lu¨thi, A., Mustovic, Psychiatry, 136, 347–353. H., Dammann, G., et al. (2003). Differential sex-inde- Rogers, M.A., Kasai, K., Koji, M., Fukuda, R., Iwanami, A., pendent amygdala response to infant crying and laugh- Nakagome, K., et al. (2004). Executive and prefrontal ing in parents versus nonparents. Biological Psychiatry, dysfunction in unipolar depression: A review of neuro- 54, 1367–1375. psychological and imaging evidence. Neuroscience Serbin, L.A., & Karp, J. (2004). The intergenerational Research, 50, 1–11. transfer of psychosocial risk: Mediators of vulnerability Rosenblatt, J.S. (1969). The development of maternal and resilience. Annual Review of Psychology, 55, 333– responsiveness in the rat. American Journal of Ortho- 363. psychiatry, 39, 36–56. Sergerie, K., Chochol, C., & Armony, J.L. (2008). The role of Rosenblatt, J.S. (1975). Prepartum and postpartum regu- the amygdala in emotional processing: A quantitative lation of maternal behaviour in the rat. Ciba Foundation meta-analysis of functional neuroimaging studies. Neu- Symposium, 33, 17–37. roscience and Biobehavioral Reviews, 32, 811–830. Rosenblatt, J.S., Olufowobi, A., & Siegel, H.I. (1998). Sewards, T.V., & Sewards, M.A. (2002). Fear and power- Effects of pregnancy hormones on maternal responsive- dominance drive motivation: Neural representations and ness, responsiveness to estrogen stimulation of maternal pathways mediating sensory and mnemonic inputs, and behavior, and the lordosis response to estrogen stimu- outputs to premotor structures. Neuroscience and Bio- lation. Hormones and Behavior, 33, 104–114. behavioral Reviews, 26, 553–579. Rosenblum, L.A., Smith, E.L.P., Altemus, M., Scharf, B.A., Sewards, T.V., & Sewards, M.A. (2003). Representations of Owens, M.J., Nemeroff, C.B., Gorman, J.M., et al. motivational drives in mesial cortex, medial thalamus, (2002). Differing concentrations of corticotrophin-releas- hypothalamus and midbrain. Brain Research Bulletin, ing factor and oxytocin in the cerebrospinal fluid of 61, 25–49. bonnet and pigtail macaques. Psychoneuroendocrinolo- Sheline, Y.I., Barch, D.M., Price, J.L., Rundle, M.M., gy, 27, 651–660. Vaishnavi, S.N., Snyder, A.Z., et al. (2009). The default Roth, T.L., Lubin, F.D., Funk, A.J., & Sweatt, J.D. (2009). mode network and self-referential processes in depres- Lasting epigenetic influence of early-life adversity on the sion. Proceedings of the National Academy of Sciences of BDNF gene. Biological Psychiatry, 65, 760–769. the United States of America, 106, 1942–1947. Rudorfer, M.V., Risby, E.D., Osman, O.T., Gold, P.W., & Shin, L.M., Rauch, S.L., & Pitman, R.K. (2006). Amygdala, Potter, W.Z. (1991). Hypothalamic–pituitary–adrenal medial prefrontal cortex, and hippocampal function in axis and monoamine transmitter activity in depression: PTSD. Annals of the New York Academy of Sciences, A pilot study of central and peripheral effects of elec- 1071, 67–79. troconvulsive therapy. Biological Psychiatry, 29, 253– Siegle, G. J., Thompson, W., Carter, C.S., Steinhauer, S.R., 264. & Thase, M.E. (2007). Increased amygdala and Sarinopoulos, I., Grupe, D.W., Mackiewicz, K.L., Herring- decreased dorsolateral prefrontal BOLD responses in ton, J.D., Lor, M., Steege, E.E., et al. (2009). Uncertainty unipolar depression: Related and independent features. during anticipation modulates neural responses to aver- Biological Psychiatry, 61, 198–209. sion in human insula and amygdala. Cerebral Cortex, 20, Simon, J.J., Walther, S., Fiebach, C.J., Friederich, H., 929–940. Stippich, C., Weisbrod, M., et al. (2010). Neural reward Savitz, J.B., & Drevets, W.C. (2009). Imaging phenotypes of processing is modulated by approach- and avoidance- major depressive disorder: Genetic correlates. Neurosci- related personality traits. NeuroImage, 49, 1868–1874. ence, 164, 300–330. Smeltzer, M.D., Curtis, J.T., Aragona, B.J., & Wang, Z. Saxe, R., Moran, J.M., Scholz, J., & Gabrieli, J. (2006). (2006). Dopamine, oxytocin, and vasopressin receptor Overlapping and non-overlapping brain regions for the- binding in the medial prefrontal cortex of monogamous ory of mind and self reflection in individual subjects. and promiscuous voles. Neuroscience Letters, 394, 146– Social Cognitive and Affective Neuroscience, 1, 229–234. 151.

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health. Neurobiological perspectives on mothering 397

Smoski, M.J., Felder, J., Bizzell, J., Green, S.R., Ernst, M., Trehub, S., Unyk, A., & Trainor, L.J. (1993). Maternal Lynch, T.R., et al. (2009). fMRI of alterations in reward singing in cross-cultural perspective. Infant Behavior selection, anticipation, and feedback in major depressive and Development, 16, 285–295. disorder. Journal of Affective Disorders, 118(1–3), 69–78. Trevathan, W.R. (1983). Maternal ‘en face’ orientation Spertus, I.L., Yehuda, R., Wong, C.M., Halligan, S., & during the first hours after birth. American Journal of Seremetis, S.V. (2003). Childhood emotional abuse and Orthopsychiatry, 53, 92–99. neglect as predictors of psychological and physical van IJzendoorn, M.H. (1992). Intergenerational transmis- symptoms in women presenting to a primary care sion of parenting: A review of studies in nonclinical practice. Child Abuse and Neglect, 27, 1247–1258. populations. Developmental Review, 12, 76–99. Spreng, R.N., & Grady, C.L. (2010). Patterns of brain van Sleuwen, B.E., Engelberts, A.C., Boere-Boonekamp, activity supporting autobiographical memory, prospec- M.M., Kuis, W., Schulpen, T.W.J., & L’Hoir, M.P. (2007). tion, and theory of mind, and their relationship to the Swaddling: A systematic review. Pediatrics, 120(4), default mode network. Journal of Cognitive Neuroscience, e1097–1106. doi: 10.1542/peds.2006-2083. 22, 1112–1123. van Veen, V., & Carter, C.S. (2002). The anterior cingulate Stallings, J., Fleming, A.S., Corter, C., Worthman, C., & as a conflict monitor: fMRI and ERP studies. Physiology Steiner, M. (2001). The effects of infant cries and odors and Behavior, 77, 477–482. on sympathy, cortisol, and autonomic responses in new Vik, T., Grote, V., Escribano, J., Socha, J., Verduci, E., mothers and nonpostpartum women. Parenting: Science Fritsch, M., et al. (2009). Infantile colic, prolonged crying and Practice, 1, 71–100. and maternal postnatal depression. Acta Paediatrica, 98, Strathearn, L., Fonagy, P., Amico, J., & Montague, P.R. 1344–1348. (2009). Adult attachment predicts maternal brain and Vogt, B.A., Nimchinsky, E.A., Vogt, L.J., & Hof, P.R. (1995). oxytocin response to infant cues. Neuropsychopharma- Human cingulate cortex: Surface features, flat maps, cology, 34, 2655–2666. and cytoarchitecture. Journal of Comparative Neurology, Strathearn, L., Li, J., Fonagy, P., & Montague, P.R. (2008). 359, 490–506. What’s in a smile? Maternal brain responses to infant Vogt, B.A., Vogt, L., Farber, N.B., & Bush, G. (2005). facial cues. Pediatrics, 122, 40–51. Architecture and neurocytology of monkey cingulate Swain, J.E. (2008). Baby stimuli and the parent brain: gyrus. Journal of Comparative Neurology, 485, 218– Functional neuroimaging of the neural substrates of 239. parent–infant attachment. Psychiatry, 5, 28–36. Wang, C., Ulbert, I., Schomer, D.L., Marinkovic, K., & Swain, J.E., Lorberbaum, J.P., Kose, S., & Strathearn, L. Halgren, E. (2005). Responses of human anterior cingu- (2007). Brain basis of early parent–infant interactions: late cortex microdomains to error detection, conflict Psychology, physiology, and in vivo functional neuroi- monitoring, stimulus–response mapping, familiarity, maging studies. Journal of Child Psychology and Psychi- and orienting. Journal of Neuroscience, 25, 604–613. atry, 48, 262–287. Williams, M.A., McGlone, F., Abbott, D.F., & Mattingley, Swain, J.E., Tasgin, E., Mayes, L.C., Feldman, R., Consta- J.B. (2005). Differential amygdala responses to happy ble, R.T., & Leckman, J.F. (2008). Maternal brain and fearful facial expressions depend on selective atten- response to own baby-cry is affected by cesarean section tion. NeuroImage, 24, 417–425. delivery. Journal of Child Psychology and Psychiatry, 49, Willinger, U., Diendorfer-Radner, G., Willnauer, R., Jo¨rgl, 1042–1052. G., & Hager, V. (2005). Parenting stress and parental Swain, J.E., & Lorberbaum, J.P. (2008). Imaging the bonding. Behavioral Medicine, 31, 63–69. human parental brain. Neurobiology of the parental Wind, T.W., & Silvern, L. (1994). Parenting and family brain (pp. 83–100). Burlington, MA: Academic Press. stress as mediators of the long-term effects of child Sweeney, J.A., Kmiec, J.A., & Kupfer, D.J. (2000). Neuro- abuse. Child Abuse and Neglect, 18, 439–453. psychologic impairments in bipolar and unipolar mood Winstanley, C.A., Theobald, D.E.H., Cardinal, R.N., & disorders on the CANTAB neurocognitive battery. Bio- Robbins, T.W. (2004). Contrasting roles of basolateral logical Psychiatry, 48, 674–684. amygdala and orbitofrontal cortex in impulsive choice. Tamis-LeMonda, C.S., & Bornstein, M H. (1989). Habitu- Journal of Neuroscience, 24, 4718–4722. ation and maternal encouragement of attention in Wise, R.A. (2000). Interactions between medial prefrontal infancy as predictors of toddler language, play, and cortex and meso-limbic components of brain reward representational competence. Child Development, 60, circuitry. Progress in Brain Research, 126, 255–262. 738–751. Xie, R., He, G., Koszycki, D., Walker, M., & Wen, S.W. Tamis-LeMonda, C.S., Bornstein, M.H., & Baumwell, L. (2009). Fetal sex, social support, and postpartum (2001). Maternal responsiveness and children’s achieve- depression. Canadian Journal of Psychiatry, 54, 750– ment of language milestones. Child Development, 72, 756. 748–767. Zahn, R., de Oliveira-Souza, R., Bramati, I., Garrido, G., & Thoman, E.B. (2006). Co-sleeping, an ancient practice: Moll, J. (2009). Subgenual cingulate activity reflects Issues of the past and present, and possibilities for the individual differences in empathic concern. Neuroscience future. Sleep Medicine Reviews, 10, 407–417. Letters, 457, 107–110. Thulier, D. (2009). Breastfeeding in America: A history of influencing factors. Journal of Human Lactation, 25, 85– Accepted for publication: 22 June 2010 94. Published online: 1 October 2010

2010 The Authors Journal of Child Psychology and Psychiatry 2010 Association for Child and Adolescent Mental Health.